Table 5. Participant characteristics by response to the statement: "Parts of the booklet are upsetting".
| Strongly Agree or Agree | Strongly Disagree, Disagree, Neither | P-valuea | Absolute effect sizea | |
|---|---|---|---|---|
| Proxy and interview characteristics | (N = 54) | (N = 68) | ||
| Age, median (IQR)b | 47 (34, 60) | 54 (46, 63) | 0.04 | 0.38 |
| Female, n (%)b | 37 (70%) | 46 (69%) | 1.00 | 0.03 |
| Years of education, median (IQR) | 14 (12, 16) | 14 (12, 16) | 0.20 | 0.10 |
| Self-identified race, n (%)b | ||||
| Black or African American | 22 (41%) | 33 (49%) | 0.50 | 0.29 |
| White | 26 (48%) | 29 (43%) | ||
| Other | 2 (4%) | 6 (9%) | ||
| Relation to Patient, n (%)c | ||||
| Spouse/Partner | 21 (39%) | 25 (37%) | 0.55 | 0.06 |
| Adult child | 17 (32%) | 23 (34%) | ||
| Parent | 6 (11%) | 8 (12%) | ||
| Other | 10 (19%) | 12 (18%) | ||
| ICU day at time of interview, median (IQR) | 2 (1, 3) | 2 (1, 3) | 0.79 | 0.04 |
| "Do you have legal authority to speak for [patient's name] if he/she is unable to make decisions about medical treatment?" This question was asked BEFORE the proxy viewed the booklet. | ||||
| Yes | 33 (61%) | 48 (71%) | 0.36 | 0.21 |
| "Have you ever supported a loved one in an ICU before?", n (%)b | ||||
| Yes | 33 (61%) | 42 (62%) | 0.89 | 0.06 |
| Patients characteristics & Outcomesd | ||||
| Age, median (IQR) | 58 (45, 69) | 58 (50, 70) | 0.41 | 0.14 |
| Female, n (%) | 25 (46%) | 35 (51%) | 0.70 | 0.10 |
| Income of zip code in $1000s of USD, median (IQR)e | 58.6 (36.0, 74.1) | 59.1 (36.0, 82.0) | 0.90 | 0.03 |
| Is the patient "full code" during the interview?f | ||||
| Yes | 43 (80%) | 60 (88%) | 0.29 | 0.32 |
| Location prior to hospitalization, n (%)b | ||||
| House/Apartment (independent) | 39 (72%) | 42 (62%) | 0.32 | 0.32 |
| House/Apartment (with assistance) | 13 (24%) | 18 (26%) | ||
| Other | 2 (4%) | 7 (10%) | ||
| Admission diagnosis, n (%) | ||||
| Respiratory failure | 27 (50%) | 29 (43%) | 0.28 | 0.35 |
| Sepsis | 10 (19%) | 14 (21%) | ||
| Gastrointestinal | 7 (13%) | 4 (6%) | ||
| Other | 10 (19%) | 21 (31%) | ||
| In-hospital death, n (%) | 15 (28%) | 19 (28%) | 1.00 | 0.00 |
Abbreviation: ICU, Intensive care unit; IQR, Interquartile Range; USD, United States Dollar
a Absolute effect size = absolute value of difference in means or proportions divided by standard error. P-values obtained from the Wilcoxon-Mann-Whitney two-sample test for continuous values, and the Chi-square test for categorical values with Fisher’s exact test for cell-sizes <10.
b Proxies declined to report age (n = 2), sex (n = 2), race (n = 4), and prior experience as an ICU proxy (n = 3). Location prior to hospitalization missing for 1 patient.
c Percentages do not sum to 100% due to rounding.
d 2 proxies were interviewed for 11 patients, creating 11 pairs of proxies independently answering questions about the same patient at different times during the ICU stay. In 4 dyads, the proxies gave discordant responses about whether the booklet was upsetting.
e US Census Bureau 2010–2014; $41,819 median household income for Baltimore City; $74,194 median household income for MD state. No zip code was provided for 1 international patient.
f 54 (53%) of patients who are herein designated as full code didn’t have any documented code status during the interview.