718

Abstract

Introduction: High quality physician-family communication in the intensive care unit (ICU) is essential in order to provide comprehensive, family-centered care. However, prior research has shown that the quality of physician-family communication in the ICU is often sub-optimal, and may be even worse for families with limited English proficiency. Most existing data in this area come from retrospective, survey based studies performed in the adult ICU setting. We performed the first prospective analysis of the quality of physician-family communication during interpreted and non-interpreted family meetings in the pediatric ICU setting by audio-recording 30 family meetings in English and Spanish. After examining our transcripts, we hypothesized that the quality of communication during interpreted meetings may actually be superior to non-interpreted meetings due to simpler physician speech, increased opportunity for family speech, and interpreter clarifications that facilitate shared decision-making. Methods: Thirty pediatric ICU family meetings (21 English and 9 Spanish) were audio-recorded, transcribed, and analyzed using the qualitative method of directed content analysis. Quality of communication was analyzed in three ways: 1) presence of elements of shared decision-making, 2) balance between physician and family speech, and 3) complexity of physician speech. Only meetings containing a decisional component were analyzed for the elements of shared decision-making. The balance between physician and family speech was measured by time spent speaking during meetings, as well as weighted percentage of total meeting time spent speaking (speech time divided by physician plus family speech time). Complexity of physician speech was determined by analyzing physician speech passages and assigning a Flesch-Kincaid grade level score. Results: Elements of shared decision making occurred infrequently during both interpreted and non-interpreted meetings. Only 4 of the 11 elements were present in over half of English meetings, and only 3 of the 11 were present in over half of Spanish meetings. Physicians spent more time speaking than families. Physicians spoke for a mean of 20.7 minutes (95% CI 16 - 25.4), while families spoke for 9.3 minutes (95% CI 5.3 - 13.4) during English meetings. During Spanish meetings, physicians spoke for a mean of 14.9 minutes (95% CI 9.7 - 20.1) versus just 3.7 minutes (95% CI 1.5 - 5.9) of family speech. By weighted percentage, English meetings contained 71% physician speech (95% CI 63% - 78%) versus 29% family speech (95% CI 22% - 37%), compared to 80% physician speech (95% CI 73% - 88%) versus 20% family speech (95% CI 12% - 27%) in Spanish meetings. Physician speech complexity received a mean grade level score of 8.2 (95% CI 7.3 - 9.1) in English meetings, compared to 7.2 (95% CI 5.8 - 8.6) in Spanish meetings. Conclusions: The quality of physician-family communication during interpreted and non-interpreted family meetings in the pediatric ICU is poor overall. Interpreted meetings had poorer communication quality as evidenced by fewer elements of shared decision-making and greater imbalance between physician and family speech. However, physician speech may be less complex during interpreted meetings versus non-interpreted meetings. Our data suggest that physicians can improve communication in both interpreted and non-interpreted family meetings by increasing the use of elements of shared decision-making, improving the balance between physician and family speech, and decreasing the complexity of physician speech.