SBIRT as a Vital Sign for Behavioral Health Identification, Diagnosis, and Referral in Community Health Care

METHODS

Using a quasi-experimental design, we undertook a prospective, continuous quality improvement study of a screening program for behavioral health by evaluating outcome data.

The research hypothesis considered whether using the SBIRT screening process would result in improved identification of medical patients who might require intervention, referral, and treatment for the key indicators of depression, alcohol abuse, and substance abuse when compared with similar patients in a control setting.

Specifically, the study setting involved the Ohio North East Health Systems (ONE Health), a FQHC operating 6 primary care facilities in northeast Ohio. ONE Health draws from a target population in which more than 90% are at or below 200% of the federal poverty level.⁶

To compare the effectiveness of implementing SBIRT in this setting, we selected 2 facilities operated by ONE Health: the Youngstown Health Center was designated as the intervention site, and the Warren Health Center was chosen as the control site. Both facilities are in urban settings and are federally designated as medically underserved and health professional service areas. The sites serve a predominantly low-income population in which uninsured Medicaid and Medicare patients make up 85.5% of the medical users at the intervention site and 86.8% at the control site.

Before SBIRT implementation, the medical and clerical staff at the Youngstown Health Center received 3 educational sessions to learn about the SBIRT concept, referral process, and screening tools, and to develop motivational interviewing skills. Training sessions were conducted with the support of the Institute for Research, Education and Training in Addictions (http://ireta.org/). Clinicians were not given a participation incentive, but were updated on data collection and effectiveness of the program. The Warren Health Center maintained its standard practices of relying on clinicians’ discretion of behavioral health matters. For referrals we established off-site behavioral health partnerships with local mental health and substance abuse clinicians. Behavior health clinicians were not given a participation incentive, but they were updated on data collection and effectiveness of the program; they were unaware, however, of the nature of any ongoing studies. The Warren Health Center maintained its standard practices of relying on clinicians’ discretion with regard to evaluation, intervention, and referrals of behavioral health matters.

During the study period (February through August 2013), 2,482 medical patients, aged 18 years and older, were screened for depression, alcohol, and substance use. The screening tool, consisting of 5 dichotomous questions adapted from the Oregon Health & Science University’s SBIRT program,⁷ was introduced as a vital sign and added to the routine intake process. Screening was self-administered. Those eliciting positive responses were given appropriate, quanitifiable standardized tests: the Drug Abuse Screening Test (DAST), Alcohol Use Disorders Identification Test (AUDIT), and the Patient Health Questionnaire (PHQ-9). Results were recorded in the electronic health record (EHR) Next Gen Automated EHR (http://www.nextgen.com) and if indicated by a standardized recommendation scale, clinicians performed a brief intervention, made a behavioral health referral, and/or initiated pharmacotherapy. All activities were recorded and documented in the EHR.