Exploring Medical Residents’ Likelihood to Perform Screening, Brief Intervention, and Referral to Treatment (SBIRT) Behaviours 30 days after a Face-to-Face SBIRT Training

¹Indiana Prevention Resource Center, Department of Applied Health Science, School of Public Health, Bloomington, Indiana, USA

²Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana, USA

³Midtown Community Mental Health Center, Indianapolis, Indiana, USA

⁴Wishard/Eskenazi Health Services, Indianapolis, Indiana, USA

*Corresponding Author:

Jon Agley, Ph.D., MPH
Indiana Prevention Resource Centre
Department of Applied Health Science
School of Public Health
501 N. Morton St., Suite 110
47401, Bloomington, Indiana, USA
Tel: 812-855-3123
E-mail: jagley@indiana.edu

Received date: October 31, 2012; Accepted date:November 26, 2012; Published date: November 28, 2012

Citation: Agley J, Gassman RA, Crabb D, Babcock D, Bartholomew J, et al. (2012) Exploring Medical Residents’ Likelihood to Perform Screening, Brief Intervention, and Referral to Treatment (SBIRT) Behaviours 30 days after a Face-to-Face SBIRT Training. J Community Med Health Educ 2:188. doi:10.4172/2161-0711.1000188

Copyright: © 2012 Agley J, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Journal of Community Medicine & Health Education

Abstract

Introduction: Even though anti-retroviral treatments for HIV/AIDS can slow down the course of the disease, there is no known cure or vaccine. Preventing the infection is the key aim in controlling the AIDS pandemic. A health education intervention was conducted among pre medical students in Misurata to improve their knowledge about HIV/ AIDS and the effectiveness of the intervention was evaluated. Objectives of study: To assess the knowledge of premedical students about HIV/AIDS and to evaluate the effectiveness of a health education intervention. Method of study: An awareness study, followed by a health education intervention on HIV/AIDS was done among 160 premedical students of Misurata, Libya for a period of 4 months. Assessment of baseline knowledge was followed by a health education intervention. Effectiveness of intervention was evaluated and improvement in post test knowledge was analyzed using t-test. Results: Knowledge about the cause of AIDS, incubation period, ability of disease to make its patient exposed to other infections, absence of complete cure and the presence of Preventive methods were excellent and knowledge about the ability to cause cancers, absence of an effective vaccine and 100% fatality were good on pretest evaluation. Knowledge about the modes of transmission of disease and the ways by which AIDS can not be transmitted were poor on pre-test evaluation. Knowledge about all aspects was excellent on post-test evaluation. Difference between pre and post test mean scores was found to be highly significant. Conclusions: The health education intervention was effective.

Keywords

SBIRT/MI; Behavioural intentions; Training sessions

Introduction

According to the National Survey of Drug Use and Health (NSDUH), which surveys Americans age 12+ living in the United States (US), approximately 131.3 million Americans were current alcohol users in 2010 [1]. A separate study found that 17.9 million Americans were estimated to have an alcohol use disorder, 89% of whom were unaware of their condition [2]. The scope of chronic illness, such as liver disease [3] and other negative health effects, which can range widely and include such things as burns, traffic accidents, and traumatic brain injuries [4], posed by alcohol use is nationally broad, potentially affecting the entire portion of the US population that has an alcohol use disorder. Further, non-communicable chronic illnesses, including those posed by alcohol use, have captured the attention of the United Nations, which held a special session on noncommunicable diseases in 2011 [5]. Importantly, excess alcohol use is one of several key behaviourally modifiable risk factors for chronic illness [6,7], meaning that the potential for efficacious prevention of alcohol use disorders is high. Screening, Brief Intervention, and Referral To Treatment (SBIRT) “is a comprehensive and integrated approach to the delivery of early intervention and treatment services through universal screening for persons with substance use disorders and those at risk,” evidence for the efficacy of which has been counting for more than two decades [8]. In the United States, both researchand service-based SBIRT projects have been supported by numerous federal agencies, including the National Institutes of Health (NIH) [9] and the Substance Abuse and Mental Health Services Administration (SAMHSA) [10]. It has received numerous national organizational endorsements, most recently from the American Psychiatric Nurses Association (APNA) [11]. There is a significant amount of research literature that likewise supports SBIRT’s efficacy in reducing alcohol consumption and frequency of at-risk drinking [12,13], providing costeffective treatment from multiple perspectives [14,15] and reducing heavy alcohol use and improving general and mental health [16].

SAMHSA began funding cooperative agreements to introduce SBIRT in medical residency programs in 2003. To date, 17 such cooperative agreements exist. Although the way in which SBIRT training can be implemented within a residency program varies by site, certain commonalities have been observed, in particular, with all 17 grantees of “incorporated Motivational Interviewing (MI) and validated screening instruments in the curriculum” [17]. While generalizing evaluation results from one grantee site to the larger population of medical residents in the United States is not possible, such findings are nonetheless informative. Initial research at one site suggests that medical residents tend to express satisfaction with their SBIRT training [18]. Researchers at another site focusing on pediatrics medical residents found that 53% of medical residents had performed at least one brief negotiated interview within 9 months of their baseline training experience [19].

Impetus for the Current Study

At Indiana University School of Medicine (IUSM), Indianapolis, first year medical residents (PGY1) who participated in the SBIRT training protocol were asked to complete an online introduction to SBIRT and then participated in a 4-6 hour face-to-face training. Each training session was opened by a statement of support from at least one medical doctor/professor of clinical medicine at IUSM. They were conducted by one or more licensed clinical social workers from Midtown Community Mental Health Centre who collaborated with IUSM to develop the training curriculum as a component of the grant project. At least one social worker at each training session was a member of the Motivational Interviewing Network of Trainers (MINT) [20].

Using the protocol described in the Methods section, project evaluators collected baseline data (immediately post-training) and 30-day follow-up data from medical residents who participated in the face-to-face training sessions. Of particular interest to the project’s steering committee were two items measuring anticipated resident behaviour regarding SBIRT/MI. These questions asked residents about their likelihood of performing desired SBIRT-related behaviours. Asking about the likelihood of performing behaviours is a frequently used method of assessing behavioural intentions [21-24] which are an immediate antecedent to behavioural performance – this is the principle that underlies many important behaviour modelling techniques, such as the Theory of Reasoned Action (TRA) [25] and the Theory of Planned Behaviour (TPB) [26]. Stated likelihood of behaviour performance significantly has been associated with actual behaviour performance [27]. Measurement of these variables therefore served as an initial indicator of training efficacy to project staff. Although evaluators reported initially high levels of behavioural intentions to perform desired SBIRT-related behaviours among medical residents, these levels declined after 30 days, and a trend between training cohorts was documented in evaluation reports submitted to the steering committee. Among the sample used in the current study, paired-sample t-tests indicate a significant decline in the mean level of likelihood, measured on a scale of 1-5, from baseline to follow-up in terms of likelihood of asking patients about alcohol consumption patterns (mA-B=-.154, df=148, t=-3.33, p=.001) and likelihood of stating medical concerns about drinking patterns to patients (mA-B=-.101, df=148, t=-2.27, p=0.025). These findings were considered problematic, not only because residents are expected to perform SBIRT/MI following their training, but also because other studies have indicated that clinically significant effects on attitudes toward behaviour change (i.e., MI) in clinical care have remained significant at a four-week follow-up point following a training session [28]. While attitudes are fundamentally different constructs than behavioural intentions, maintenance of any affective or intentional by products of SBIRT training for medical residents in one or more cases posits that such a goal is not unreasonable for our program.

Therefore, this exploratory study used binary logistic regression models and frequency data to investigate factors, including race, gender, self-efficacy, endorsement of SBIRT’s efficacy, comfort with the topic area, and expected benefits of the training for clients, that might potentially predict reporting a decline between baseline and follow-up in two likelihood variables. The information produced by these analyses was used to inform USM’s SBIRT in Medical Residency program, and dissemination of these results is intended to benefit current and future cohorts of medical residents in all locations who might receive SBIRT/ MI instruction.

Methods

Participants and procedures

This study assessed all medical residents in PGY1 (first year of residency) in the internal medicine, pediatrics, family medicine, and obstetrics/gynaecology programs at Indiana University School of Medicine, Indianapolis, who completed a face-to-face SBIRT/MI training between April, 2010 and August, 2012. The institutional review board at Indiana University approved the research protocol prior to survey administration. Program evaluators or project managers from the school’s SBIRT in Medical Residency program administered a baseline survey immediately following each of 28 trainings that were attended by a total of 191 residents. The survey instrument included a passive, IRB-approved consent letter and a brief set of instructions for completion. All residents who were asked to complete the survey at baseline did so (100% response rate). Residents’ e-mail addresses were collected on the baseline instrument in order to facilitate 30-day followup data collection. Thirty days following each resident’s completion of the baseline instrument, he/she was sent an e-mail from the program evaluator, signed by the chairman of the Department of Medicine, inviting him/her to participate in a follow-up survey administered online through Survey Monkey. Each week for the subsequent three weeks, non-responding residents received a follow-up e-mail to encourage participation. In total, 149 residents completed the followup survey instrument in full (78.0% response rate). No significant socio demographic differences (gender or race) were observed between respondents and non-respondents at follow-up.

Instrument

For both the baseline and follow-up instruments, the first half of the survey was comprised of SAMHSA’s Customer Service Training (CST) questionnaire, administered as a grantee requirement per the Government Performance and Results Act (GPRA). Additional research questions comprised the second half of both the baseline and follow-up surveys. The final baseline survey contained 30 closed-ended questions, and the final follow-up survey contained 22 closed-ended questions.

Not all items asked on the surveys were used in this study because we particularly were interested in residents’ individual characteristics rather than their perceptions of the training. Table 1 contains a list of all variables used in these analyses. On the baseline survey, we measured self-efficacy (“I am currently effective when working in this topic area”) and expected benefits of the training (“I expect this training to benefit my clients”) using items included on the CST questionnaire. These items were measured on a scale of 1-5 (1: Strongly agree, 5: Strongly disagree). We likewise used CST questionnaire items to assess gender and race.

Table 1: Characteristics of the sample at baseline (n=149).

We measured residents’ comfort with the topic area of SBIRT (substance use) with the question “I am comfortable asking about a patient’s substance use patterns,” and we measured residents’ endorsement of SBIRT’s efficacy using the question “Given adequate information and training, primary care practitioners can help patients reduce their alcohol consumption.” These items were measured on a scale of 1-5 (1: Strongly agree, 5: Strongly disagree). We measured residents self-reported likelihood of performing desired outcome behaviours related to SBIRT within 12 months using two questions previously established in the scientific SBIRT literature [22]: (a) “On average, how likely is it in the next 12 months that you will ask patients who are current drinkers about their alcohol consumption patterns?” (b) “On average, how likely is it in the next 12 months that you will state to patients your medical concerns about their drinking patterns or related health risks?” These items were measured on a scale of 1-5 (1: Very likely, 5: Very unlikely). All questions were checked for content validity by a panel of experts (n=7) including medical doctors, social workers, registered nurses, and health behaviour researchers.

Analysis

We used SPSS version 20.0 to conduct the statistical analyses in this study. We articulated two binomial logistic regression models using the 137 medical residents who fully completed the baseline survey and answered both likelihood questions on the follow-up survey. The predictor variables in these models were race, gender, selfefficacy, expected benefits of SBIRT, comfort with substance use, and endorsement of SBIRT (Table 1) as measured on the baseline survey. These variables each were dichotomized by identifying the integer response option that most closely separated the respondents into two equal groups (i.e., was closest to the median value) in order to avoid over specifying the model [29] which would result in individual cases becoming identifiable by specific combinations of responses.

The outcome variables in these models were likelihood variables. In order to determine which individuals were “less likely” to ask current drinkers about their alcohol consumption patterns or to state medical concerns about (patients) drinking patterns or related health risks 30 days after the baseline training versus baseline, we:

(1) Determined which residents were equally or more likely to perform the given behaviours at follow-up vs. baseline and assigned these individuals a value of 0.

(2) Determined which residents were less likely to perform the given behaviours at follow-up vs. baseline and assigned these individuals a value of 1.

Each of the two likelihood variables separately was assessed by a logistic regression model. In both cases, predictor variables were checked for collinearity using Pearson correlation analysis. No issues with the data were observed. Hosmer-Lemeshow tests [30] were nonsignificant, indicating that each of the models fit the survey data. All odds ratios reported in this study are adjusted odds ratios, meaning that they account for the predictive value of a variable only in the context of the whole model.

Results

Descriptive findings

As shown in table 1, the majority of the respondents were white (70%), and slightly more than half were male (54%). At baseline, 60% of respondents strongly agreed or agreed that they were currently effective in the topic area of SBIRT/MI, 44% strongly agreed that they were comfortable asking about a patient’s substance use patterns, 60% strongly agreed that primary care practitioners can help patients reduce their alcohol consumption, and 59% strongly agreed that the SBIRT/MI training would benefit their clients. In comparing baseline to follow-up data, we found that 24% of respondents were less likely to ask patients who are current drinkers about their alcohol consumption at follow-up versus at baseline, and 20% of respondents were less likely to state their medical concerns to patients about their drinking patterns and related health risks at follow-up versus at baseline.

Multivariate analyses

As shown in table 2, one variable or race, emerged as a significant predictor of ‘decreased likelihood to state medical concerns to patients about their drinking patterns or related health risks.’ In particular, non-white respondents were more likely than white respondents to report decreased likelihood to state medical concerns to patients about their drinking patterns or related health risks at follow-up versus at baseline (adjusted OR=2.63, p=0.35). While no variables emerged as significant predictor variables in the first model (‘decreased likelihood to ask about alcohol consumption patterns’), one variable, gender, approached statistical significance (adjusted OR=2.28, p=.066). We mention this here because the estimated effect size [31] calculated as (ln[2.28]/1.81), was moderate (0.46).

Table 2: Adjusted odds of reporting decreased likelihood of two target behaviors.

Discussion

Our findings primarily focus on logistic regression analyses examining two likelihood outcomes related to asking patients about their alcohol consumption patterns and stating medical concerns about their drinking patterns or related health risks. As noted elsewhere in this paper, it is important that intentions to perform desired behaviours expressed at baseline (immediately following training) are retained longitudinally in order to train to demonstrate a meaningful impact on physician behaviour. We found that non-white physicians are significantly more likely than white physicians to report decreased likelihood of stating medical concerns about patients’ drinking patterns or related health risks 30 days following their face-to-face training compared to a baseline measure. This exploratory finding has the potential to be meaningful both in terms of SBIRT/MI and in terms of physician education as a broader field.

While studies examining physicians’ race form a significant part of the current peer-reviewed medical literature, the preponderance examine the issue of “race-concordance,” a sub-field of study which assesses the various effects on and perceptions of patient care that occur as a result of matching physicians and patients of similar races (i.e., an African-American patient with an African-American doctor) [32,33]. Race-concordance, though, doesn’t explain the findings in this study regarding the likelihood that non-white physicians will perform specific preventive behaviours in this study because the measures of likelihood were independent of patient encounters. One possible explanation for our finding lies in the concept of “normative perceptions,” a component of the TRA [25] and TPB [26]. Non-white physicians, especially those whose graduate medical experience was obtained outside of the United States, may have had instruction that places significant emphasis on physical examinations [34]. These physicians may view stating medical concerns about drinking questions as non-normative, meaning they may not see their role as physicians as involving conversations with patients about their drinking patterns and related health risks. Another study of preventive screening behaviours among physicians, specifically suicide risk assessments, also found that non-white physicians were less likely to perform desired screening behaviours than white physicians [35]. In that study, the authors speculate that physicians’ values and beliefs might influence their intentions to perform screening behaviours. If preventive screening is not understood as “what physicians do,” or if physicians do not believe that positive medical outcomes can accrue from the use of a screening tool coupled with a brief intervention, then some physicians may be less likely to incorporate SBIRT/MI processes into their practice.

Our findings in this area, in conjunction with those also cited herein, suggest the need to further emphasize the extent to which alcohol consumption is a highly relevant component of patients’ medical care during SBIRT/MI training sessions. It may also be helpful to remind residents of the wide variety of scholarly and professional organizations that have endorsed SBIRT/MI practices as a component of primary care. SBIRT/MI and the integration of behavioural health and primary care is currently a priority issue for public health in the United States [36]. Further, additional research should not only work to confirm the generalizability of our exploratory finding, but also should seek to measure physicians’ normative perceptions about preventive screening, both for alcohol and other drug use, and for other key behavioural health risks, to test the explanation proposed herein.

Although gender was not a significant predictor variable in our regression models, it approached statistical significance in one model and demonstrated a moderate effect size. Given the size of the sample (n=137) and the effect size, it is possible that a replication of this study among a larger number of medical residents would yield significant results. A significant result would indicate that males are less likely than females to retain likelihood to ask patients who are current drinkers (in this context, who ‘pre screened’ positive for a potential alcohol use disorder) about their current drinking patterns 30 days after their training versus immediately following the training. This, in turn, would not be surprising – research literature indicates that female physicians are more likely than male physicians to provide preventive services, which, in the context of this study, would include inquiry into alcohol consumption patterns, to their patients [37]. This study has several strengths. Notably, there are only a moderate number of medical residency SBIRT training programs in the United States – the ability to capture research data from such a training program is still limited. In addition, the TRA and TPB are highly validated health behaviour theories, and they function as a strong underpinning for components of our survey instrument. However, this study has several limitations as well. The sample size is sufficiently small that it restricted the analyses’ power. Further, because all of the participants in this study were from the same medical residency program, these results cannot be generalized to a population outside of that program, although, as first year medical residents, the participants originated from a variety of different medical schools.

Nonetheless, the exploratory results reported in this study have the potential to improve the way in which SBIRT/MI trainings are offered during medical residency and, in turn, to increase the number of residents who retain intentions to perform SBIRT-related services longitudinally.

References

1. http://www.samhsa.gov/data/NSDUH/2k10NSDUH/2k10Results.htm
2. Clark HW, Power AK, Le Fauve CE, Lopez EI (2008) Policy and practice implications of epidemiological surveys on co-occurring mental and substance use disorders. J Subst Abuse Treat 34: 3-13.
3. Szabo G (2010) The 40th anniversary of the National Institute on Alcoholism and Alcohol Abuse: The impact on liver disease. Hepatology 52: 10-12.
4. Levy DT, Mallonee S, Miller TR, Smith GS, Spicer RS, et al. (2004) Alcohol involvement in burn, submersion, spinal cord, and brain injuries. Med Sci Monit 10: CR17-24.
5. Room R, Rehm J, Parry C (2011) Alcohol and non-communicable diseases (NCDs): Time for a serious international public health effort. Addiction 106: 1547-1548.
6. Anstiss T (2009) Motivational interviewing in primary care. J Clin Psychol Med Settings 16: 87-93.
7. Mokdad AH, Marks JS, Stroup DF, Gerberding JL (2004) Actual causes of death in the United States, 2000. JAMA 291: 1238-1245.
8. Babor TF, McRee BG, Kassebaum PA, Grimaldi PL, Ahmed K, et al. (2007) Screening, brief intervention, and referral to treatment (SBIRT): Toward a public health approach to the management of substance abuse. Subst Abus 28: 7-30.
9. http://officeofbudget.od.nih.gov/index.htm
10. http://www.samhsa.gov/prevention/sbirt/
11. http://www.prweb.com/releases/2012/SBIRTPositionPaper/prweb9943525.htm
12. Bertholet N, Daeppen JB, Wietlisbach V, Fleming M, Burnand B (2005) Reduction of alcohol consumption by brief alcohol intervention in primary care: Systematic review and meta-analysis. Arch Intern Med 165: 986-995.
13. Cherpitel CJ, Moskalewicz J, Swiatkiewicz G, Ye Y, Bond J (2009) 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 70: 982-990.
14. Solberg LI, Maciosek MV, Edwards NM (2008) Primary care intervention to reduce alcohol misuse ranking its health impact and cost effectiveness. Am J Prev Med 34: 143-152.
15. Quanbeck A, Lang K, Enami K, Brown RL (2010) A cost-benefit analysis of Wisconsin's screening, brief intervention, and referral to treatment program: adding the employer's perspective. WMJ 109: 9-14.
16. Madras BK, Compton WM, Avula D, Stegbauer T, Stein JB, et al. (2009) Screening, brief interventions, referral to treatment (SBIRT) for illicit drug and alcohol use at multiple healthcare sites: Comparison at intake and six months. Drug Alcohol Depend 99: 280-295.
17. . Pringle JL, Kowalchuk A, Meyers JA, Seale JP (2012) Equipping residents to address alcohol and drug abuse: The National SBIRT Residency Training Project. JGME 4: 58-63.
18. Bray JH, Kowalchuk A, Waters V, Laufman L, Shilling EH (2012) Baylor SBIRT Medical Residency Training Program: Model description and initial evaluation. Subst Abus 33: 231- 240.
19. Ryan SA, Martel S, Pantalon M, Martino S, Tetrault J, et al. (2012) Screening, brief intervention, and referral to treatment (SBIRT) for alcohol and other drug use among adolescents: Evaluation of a pediatric residency curriculum. Subst Abus 33: 251-260.
20. http://www.motivationalinterviewing.org/
21. Rothman AJ, Salovey P, Antone C, Keough K, Martin CD (1993) The influence of message framing on intentions to perform health behaviors. J Exp Soc Psychol 29: 408-433.
22. Gassman RA (2003) Medical specialization, profession, and mediating beliefs that predict stated likelihood of alcohol screening and brief intervention: Targeting educational interventions. Subst Abus 24: 141-156.
23. Beale DA, Manstead ASR (1991) Predicting mothers’ intentions to limit frequency of infants’ sugar intake: Testing the Theory of Planned Behavior. J Appl Soc Psychol 21: 409- 431.
24. Lam T, Hsu CHC (2006) Predicting behavioral intention of choosing a travel destination. Tourism Manage 27: 589-599
25. Azjen I, Fishbein M (1980) Understanding attitudes and predicting social behavior. Prentice- Hall, Englewood, NJ.
26. Azjen I (1991) The Theory of Planned Behavior. Organ Behav Hum Dec 50: 179-211.
27. Shepperd BH, Hartwick J, Warshaw PR (1988) The theory of reasoned action: A meta-analysis of past research with recommendations for modifications and future research. J Consum Res 15: 325-343.
28. Wiley EJ, Morrow D, Irwin JD (2011) The impact of a one-day applied training in motivational interviewing on health practitioners' perceived competence, autonomy, efficacy, and attitudes to facilitate behavior change: A pilot study. JCMHE.
29. Gunst RF, Mason RL (1980) Regression analysis and its application. Marcel Decker, Inc, New York.
30. Hosmer DW, Lemeshow S (2000) Applied Logistic Regression. John Wiley & Sons, Inc., New York.
31. Chinn S (2000) A simple method for converting an odds ratio to effect size for use in meta analysis. Stat Med 19: 3127-3131.
32. . Kumar D, Schlundt DG, Wallston KA (2009) Patient-physician race concordance and its relationship to perceived health outcomes. Ethn Dis 19: 345-351.
33. LaVeist TA, Nuru-Jeter A (2002) Is doctor-patient race concordance associated with greater satisfaction with care? J Health Soc Behav 43: 296-306.
34. Hooper LM, Epstein SA, Weinfurt KP, DeCoster J, Qu L, et al. (2012) Predictors of primary care physicians’ self-reported intention to conduct suicide risk assessments. J Behav Health Serv Res: 103-115.
35. Reichert SF, Thomas AK (2010) Mentoring a vulnerable cohort for success: The special case of the foreign medical graduate. In: Mentoring and academic medicine (Holly Humphrey, ed.), Versa Press.
36. Chowdhury FM, Kulcsar M, Gilchrist S, Hawkins NA (2012) News from the CDC: Integrating behavioural health into the patient-cantered medical home. Transl Behav Med 2: 257-259.
37. Bertakis KD (2009) The influence of gender on the doctor-patient interaction. Patient Educ Couns 76: 356-360.