We conducted annual telephone interviews with participants or their proxies regarding recent emergency room visits, hospitalizations, or death.
Most promising approaches include promoting explicit communication between older adults and proxies regarding future medical care, and encouraging patients to document their preferences.
Furthermore, when comparing the relationship of the proxies regarding the patients we found no significant differences when the ALDS item bank was used.
When comparing the proxies regarding gender (P = 0.251), relationship with the patient (spouse vs. child; P = 0.889) and age (< 65 years vs. ≥65 years; P = 0.904), we found no significant differences.
The level of agreement between patients and their proxies regarding the ALDS both in ICU and CCU patients was assessed with the Bland-Altman method and visually with a Bland-Altman plot [ 20].
The intra-class correlation coefficients were used to estimate the absolute agreement levels between patients and proxies regarding the ALDS as well as the PCS and MCS of the SF-12.
The questionnaire was validated by comparison of 1564 pairs of subject and proxy answers regarding frequency of alcohol consumption, which found a weighted Cohen's kappa coefficient of 0.61, indicating moderate agreement [ 21].
Analyses including all cases, as well as analyses only including the non-fatal cases, were performed, due to the fact that we had to rely on proxy information regarding the exposure of BMI and leisure-time physical activity for the fatal cases.
The OpinionFamily questionnaire, built with 6 categories (4 items each), aimed to measure the proxies' satisfaction regarding their perception of the quality of care.
