Systematic implementation of an advance directive program in nursing homes- A randomized controlled trial

Context Although advance directives are commonly used in the community, lit tle is known about the effects of their systematic implementation. Objectives To examine the effect of systematically implementing an advance directive in nursing homes on patient and family satisfaction with involvem ent in decision making and on health care costs. Design Randomized controlled trial conducted June 1, 1994, to August 31, 19 98. Setting and Participants A total of 1292 residents in 6 Ontario nursing hom es with more than 100 residents each. Intervention The Let Me Decide advance directive program included educating staff in local hospitals and nursing homes, residents, and families about advance directives and offering competent residents or next-of-kin of menta lly incompetent residents an advance directive that provided a range of hea lth care choices for life-threatening illness, cardiac arrest, and nutritio n. The 6 nursing homes were pair-matched on key characteristics, and 1 home per pair was randomized to take part in the program. Control nursing homes continued with prior policies concerning advance directives. Main Outcome Measures Residents' and families' satisfaction with health car e and health care services utilization over 18 months, compared between int ervention and control nursing homes. Results Of 527 participating residents in intervention nursing homes, 49% o f competent residents and 78% of families of incompetent residents complete d advance directives. Satisfaction was not significantly different in inter vention and control nursing homes. The mean difference (scale, 1-7) between intervention and control homes was -0.16 (95% confidence interval [CI], -0 .41 to 0.10) for competent residents and 0.07 (95% CI, -0.08 to 0.23) for f amilies of incompetent residents. intervention nursing homes reported fewer hospitalizations per resident (mean, 0.27 vs 0.48; P =.001) and less resou rce use (average total cost per patient, Can $3490 vs Can $5239; P =.01) th an control nursing homes. Proportion of deaths in intervention (24%) and co ntrol (28%) nursing homes were similar (P =.20). Conclusion Our data suggest that systematic implementation of a program to increase use of advance directives reduces health care services utilization without affecting satisfaction or mortality.