• Alexia M Torke, Patrick Monahan, Christopher M Callahan, Paul R Helft, Greg A Sachs, Lucia D Wocial, James E Slaven, Kianna Montz, Lev Inger, and Emily S Burke.
• Indiana University (IU) Center for Aging Research, Regenstrief Institute, Inc., Indianapolis, Indiana, USA; IU Division of General Internal Medicine and Geriatrics, Indianapolis, Indiana, USA; Fairbanks Center for Medical Ethics, IU Health, Indianapolis, Indiana, USA; Daniel F. Evans Center for Spiritual and Religious Values in Healthcare, IU Health, Indianapolis, Indiana, USA. Electronic address: atorke@iu.edu.
• J Pain Symptom Manage. 2017 Jan 1; 53 (1): 96-108.e4.

ContextAlthough many family members who make surrogate decisions report problems with communication, there is no validated instrument to accurately measure surrogate/clinician communication for older adults in the acute hospital setting.ObjectivesThe objective of this study was to validate a survey of surrogate-rated communication quality in the hospital that would be useful to clinicians, researchers, and health systems.MethodsAfter expert review and cognitive interviewing (n = 10 surrogates), we enrolled 350 surrogates (250 development sample and 100 validation sample) of hospitalized adults aged 65 years and older from three hospitals in one metropolitan area. The communication survey and a measure of decision quality were administered within hospital days 3 and 10. Mental health and satisfaction measures were administered six to eight weeks later.ResultsFactor analysis showed support for both one-factor (Total Communication) and two-factor models (Information and Emotional Support). Item reduction led to a final 30-item scale. For the validation sample, internal reliability (Cronbach's alpha) was 0.96 (total), 0.94 (Information), and 0.90 (Emotional Support). Confirmatory factor analysis fit statistics were adequate (one-factor model, comparative fit index = 0.981, root mean square error of approximation = 0.62, weighted root mean square residual = 1.011; two-factor model comparative fit index = 0.984, root mean square error of approximation = 0.055, weighted root mean square residual = 0.930). Total score and subscales showed significant associations with the Decision Conflict Scale (Pearson correlation -0.43, P < 0.001 for total score). Emotional Support was associated with improved mental health outcomes at six to eight weeks, such as anxiety (-0.19 P < 0.001), and Information was associated with satisfaction with the hospital stay (0.49, P < 0.001).ConclusionThe survey shows high reliability and validity in measuring communication experiences for hospital surrogates. The scale has promise for measurement of communication quality and is predictive of important outcomes, such as surrogate satisfaction and well-being.Copyright © 2016 American Academy of Hospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

15 keywords...

hide…