Table 4.
Study hypotheses, measures, and analytic strategies
| Purpose | Variables | Analytic strategy |
|---|---|---|
|
First aim: Main study outcomes |
Treatment condition and: |
Logistic regression |
| Hypothesis 1 |
· Enrollment in quitline by 6 months |
|
| Test the effects of warm handoff on quitline enrollment |
|
|
| Hypothesis 2 |
Treatment condition and: |
Logistic regression |
| Test the effects of warm handoff on post-discharge guideline-based counseling adherence |
· Completing at least 3 quitline counseling sessions (binary) |
|
| Hypothesis 3 |
Treatment condition and: |
Logistic regression |
| Test the effects of warm handoff on 6-month abstinence outcomes |
· 30-day abstinence |
|
| · 7-day abstinence | ||
| · Prolonged abstinence | ||
|
Second Aim: Mediation analyses |
Treatment condition and 1-month: |
Combined Poisson and logistic structural equation modeling |
| Hypothesis 4 |
· Satisfaction with care transition |
|
| Effect of warm handoff on care transitions, enrollment, adherence, and pharmacotherapy utilization |
· Enrollment in quitline |
|
| · No. of quitline sessions completed | ||
| · Pharmacotherapy use | ||
|
Third aim: Costs and cost-effectiveness |
Treatment condition and: |
-Average cost/arm |
| Hypothesis 5 |
· 30-day abstinence |
-Incremental cost/quit |
| Warm handoff will be more costly, but also more cost-effective, than fax referrals |
· Fixed/Variable costs |
|
| · Provider costs |
-Providers |
|
| · Participant costs |
-Participants |
|
| |
-Combined |
|
| Feasibility, strengths/weaknesses of protocols | Semistructured interviews | Qualitative analysis |