A before-after implementation trial of smoking cessation guidelines in hospitalized veterans

Background: Although most hospitalized smokers receive some form of cessation counseling during hospitalization, few receive outpatient cessation counseling and/or pharmacotherapy following discharge, which are key factors associated with long-term cessation. US Department of Veterans Affairs (VA) hospitals are challenged to find resources to implement and maintain the kind of high intensity cessation programs that have been shown to be effective in research studies. Few studies have applied the Chronic Care Model (CCM) to improve inpatient smoking cessation. Specific objectives: The primary objective of this protocol is to determine the effect of a nurse-initiated intervention, which couples low-intensity inpatient counseling with sustained proactive telephone counseling, on smoking abstinence in hospitalized patients. Key secondary aims are to determine the impact of the intervention on staff nurses' attitudes toward providing smoking cessation counseling; to identify barriers and facilitators to implementation of smoking cessation guidelines in VA hospitals; and to determine the short-term cost-effectiveness of implementing the intervention. Design: Pre-post study design in four VA hospitals Participants: Hospitalized patients, aged 18 or older, who smoke at least one cigarette per day. Intervention: The intervention will include: nurse training in delivery of bedside cessation counseling, electronic medical record tools (to streamline nursing assessment and documentation, to facilitate prescription of pharmacotherapy), computerized referral of motivated inpatients for proactive telephone counseling, and use of internal nursing facilitators to provide coaching to staff nurses practicing in non-critical care inpatient units. Outcomes: The primary endpoint is seven-day point prevalence abstinence at six months following hospital admission and prolonged abstinence after a one-month grace period. To compare abstinence rates during the intervention and baseline periods, we will use random effects logistic regression models, which take the clustered nature of the data within nurses and hospitals into account. We will assess attitudes of staff nurses toward cessation counseling by questionnaire and will identify barriers and facilitators to implementation by using clinician focus groups. To determine the short-term incremental cost per quitter from the perspective of the VA health care system, we will calculate cessation-related costs incurred during the initial hospitalization and six-month follow-up period.