Table 2. Demographic Characteristics of the Surveyed Population in Data Submitted to the Food and Drug Administration by Trials Granted an Exception From Informed Consent.
| Characteristic | No. (%) | ||
|---|---|---|---|
| Random Sampling (n = 13 997)a | Convenience Sampling (n = 28 451) | Total (n = 42 448) | |
| Sex | |||
| Data available | |||
| Unweighted data, No. | 3021 | 24 182 | 27 203 |
| Male | 1086 (35.9) | 9971 (41.2) | 11 057 (40.6) |
| Female | 1935 (64.1) | 14 211 (58.8) | 16 146 (59.4) |
| Weighted data, No. | 9971 | 0 | 9971 |
| Male | 4875 (48.9) | 0 | 4875 (48.9) |
| Female | 5096 (51.1) | 0 | 5096 (51.1) |
| All data, No. | 12 992 | 24 182 | 37 174 |
| Male | 5961 (45.9) | 9971 (41.2) | 15 932 (42.9) |
| Female | 7031 (54.1) | 14 211 (58.8) | 21 242 (57.1) |
| Data unavailable, No. | 1005 | 4269 | 5274 |
| Queried, no response | 0 | 82 (1.9) | 82 (1.6) |
| Queried, not provided | 505 (50.2) | 161 (3.8) | 666 (12.6) |
| Not queried | 0 | 0 | 0 |
| Not provided, unclear if queried | 500 (49.8) | 4026 (94.3) | 4526 (85.8) |
| Raceb | |||
| Data available, No. | 9216 | 24 312 | 33 528 |
| White | 7393 (80.2) | 17 396 (71.6) | 24 789 (73.9) |
| African American | 1267 (13.7) | 4246 (17.5) | 5513 (16.4) |
| Other | 556 (6.0) | 2670 (11.0) | 3226 (9.6) |
| Data unavailable, No. | 4802 | 4144 | 8946 |
| Refused or not known | 245 (5.1) | 210 (5.1) | 455 (5.1) |
| Not queried | 1493 (31.1) | 134 (3.2) | 1627 (18.2) |
| Queried, not provided | 1005 (20.9) | 337 (8.1) | 1342 (15.0) |
| Not provided, unclear if queried | 2059 (42.9) | 3463 (83.6) | 5522 (61.7) |
| Survey Location | |||
| United States | 12 343 (88.2) | 28 421 (99.9) | 40 764 (96.0) |
| Canada | 1654 (11.8) | 30 (0.1) | 1684 (4.0) |
Three trials contained survey data from random-digit dialing that were aggregated with the data from convenience sampling and not provided separately: 1003 by random sampling out of 6956 surveyed in 1 trial49; 1592 by random sampling out of 7453 surveyed in another50; 750 by random sampling out of 7228 surveyed in the third.60 Because the data could not be disaggregated and most individuals were queried by convenience sampling, the data from these 3 trials were grouped under the category of convenience sampling.
No data were weighted by race. In 3 trials, investigators provided the sex and race (by percentage) of individuals who participated in community consultation,49,50,60 and it is not clear if these percentages refer to those who completed surveys or all who participated in community consultation. For the purposes of analysis, we have used these percentages as reflecting those who completed surveys. In 6 surveys (conducted in 4 different trials), individuals could select more than 1 race, and investigators reported the aggregated data of all races selected: (1) 100 surveyed, 102 included44; (2) 322 surveyed, 327 included47; (3) 502 surveyed, 504 included47; (4) 400 surveyed, 410 included57; (5) 400 surveyed, 406 included57; (6) 400 surveyed, 401 included.59 As a result, the number of individuals categorized by race (14 018 individuals by random sampling and 28 456 by convenience sampling for a total of 42 474) is greater than the total number of individuals surveyed (13 997 individuals by random sampling and 28 451 by convenience sampling for a total of 42 448).