Table 3.
Secondary Trial Outcomes at 4-Week Follow-up, Intent-to-Treat Analysis
| Outcome | Text-Only Warnings (n = 1078) | Pictorial Warnings (n = 1071) | Difference (95% CI) | P Value | ||
|---|---|---|---|---|---|---|
| No. | Mean (SD) | No. | Mean (SD) | |||
| Psychosocial outcomesa | ||||||
| Thinking about warning messageb | 1002 | 2.3 (1.2) | 995 | 3.0 (1.1) | 0.63 (0.53 to 0.73) | <.001 |
| Fear elicited by warning | 1078 | 1.8 (1.1) | 1070 | 2.4 (1.3) | 0.63 (0.53 to 0.74) | <.001 |
| Negative affect | 1078 | 1.8 (1.0) | 1070 | 2.4 (1.2) | 0.64 (0.55 to 0.73) | <.001 |
| Thinking about the harms of smokingb | 1003 | 2.9 (1.1) | 995 | 3.1 (1.1) | 0.14 (0.04 to 0.24) | .01 |
| Perceived likelihood of harm from smoking | 1077 | 3.4 (0.9) | 1071 | 3.4 (0.9) | 0.03 (−0.05 to 0.10) | .49 |
| Positive smoking reinforcement attitudes | 1078 | 3.0 (1.2) | 1070 | 3.0 (1.1) | −0.02 (−0.12 to 0.08) | .71 |
| Negative smoking reinforcement attitudes | 1078 | 3.7 (1.0) | 1070 | 3.7 (1.0) | −0.03 (−0.12 to 0.05) | .48 |
| Quit intentionsa | 1078 | 2.5 (1.1) | 1071 | 2.7 (1.1) | 0.17 (0.08 to 0.26) | <.001 |
| Behavioral outcomes | ||||||
| No. of conversations about warning in past week | 1077 | 1.2 (2.1) | 1071 | 1.8 (2.5) | 0.63 (0.43 to 0.83) | <.001 |
| No. of conversations about the health risks of smoking in past week | 1077 | 1.2 (2.0) | 1071 | 1.6 (2.4) | 0.40 (0.22 to 0.59) | <.001 |
| No. of conversations about quitting smoking in past week | 1077 | 1.4 (2.1) | 1071 | 1.6 (2.4) | 0.26 (0.07 to 0.45) | .01 |
| No. of times forgoing a cigarette in past week | 1077 | 2.5 (2.8) | 1070 | 2.7 (2.9) | 0.25 (0.01 to 0.49) | .04 |
| Quit smoking for ≥7 days, % (SE) | 1078 | 3.8 (0.6) | 1071 | 5.7 (0.7) | 1.9 (0.1 to 3.7) | .04 |
a
Response scale for all psychosocial outcomes except quit intentions ranged from 1 to 5, with 5 indicating higher quantity or stronger endorsement; response scale for quit intentions ranged from 1 to 4, with 4 indicating highest intentions.
b
Thinking about warning message and thinking about the harms of smoking were not assessed until week 1 follow-up, resulting in more missing data than for other variables.