3084 — Results From a Randomized Controlled Trial to Facilitate the Use of Psychiatric Advance Directives: Lessons Learned

Strauss JL (Durham VA Medical Center), Henderson RC (Bronx VA Medical Center), Stechuchak KM (Durham VA Medical Center), Olsen MK (Durham VA Medical Center), Zervakis JB (Durham VA Medical Center), Swanson J (Duke University Medical Center), Swartz MS (Duke University Medical Center), Oddone EZ (Durham VA Medical Center), Weinberger M (Durham VA Medical Center), Butterfield MI (Durham VA Medical Center)

Objectives:
Psychiatric advance directives (PADs) are legal documents that allow competent persons to declare treatment preferences in advance of a mental health crisis, when they may lose capacity to make reliable health care decisions. PADs are consistent with the Patient Self-Determination Act and have been adopted by 25 states. VA does not have a specific policy for PADs, and their effects on veterans’ care and outcomes are unknown.

Methods:
Veterans with severe mental illness (SMI) hospitalized in the Psychiatry Service were enrolled in a randomized controlled trial to compare an intervention to increase PAD use to usual care (N = 240). Those in the intervention group were offered the opportunity to meet with a clinician and prepare a facilitated PAD that was entered into the medical record (n = 103). Participants completed assessments at 1, 6, and 12 months post-enrollment.

Results:
The intervention group did not differ from usual care on primary study outcomes: rehospitalization rates, satisfaction with care, perceived coercion, treatment motivation, working alliance, and clinical outcomes. Analyses of PAD content indicated that veterans can clearly specify treatment preferences that may be clinically useful. At 12-months post-enrollment, 77.4% who completed a PAD reported being extremely or moderately satisfied. In 9.3% of rehospitalizations at our facility during the 12-month follow up, patients who consented to voluntary admission in their PAD were involuntarily committed.

Implications:
The facilitated PAD intervention did not prove efficacious in guiding treatment or improving outcomes for veterans with SMI. Although veterans reported enthusiasm for and satisfaction with PADs, our data suggest that PADs may not have been routinely reviewed and followed to guide rehospitalization decisions. Our findings contradict a recent study of facilitated PADs in a community sample, which found that PADs decreased incidence of mental health crises. Potential patient and system variables that may have influenced our findings include: inpatient setting, diagnostic/demographic composition of sample, lack of patients’ primary clinician involvement in preparation of PADs, clinicians’ unawareness of PADs.

Impacts:
This study highlights the distinction between two separate aspects of PAD implementation: completion versus use. Strategies to change provider behavior with the goal of increased PAD use will be discussed.