Table 3.
Effectiveness of engagement promoting strategies in experimental arms.
| Tailoring | ||
|---|---|---|
| Author | EPS sub-type(s) | Significant |
| Tensil et al. 2013 | 1. Abstinence status 2. Goals/motivation to quit | Yes |
| Strecher et al. 2008 | 1. Goals/motivation to quit (depth outcome expectations condition) 2. Health + lifestyle (depth outcome expectations condition) 3. Self-efficacy + barriers to quittinga 4. Testimonials/success stories 5.Personalised sourcea | Yes |
| McClure et al. 2013 | 1. Testimonials/success stories 2. Self-efficacy + barriers to quitting | No |
| Elfeddali et al. 2012b | 1. Self-efficacy + barriers to quitting 2. Mood/negative affect 3. Level of planning | Unclear |
| Danaher et al. 2013c | 1.Readiness to quit 2. Interests | Unclear |
| Houston et al. 2015c | 1. Readiness to quit | Unclear |
| Severson et al. 2008c | 1. Abstinence status 2. Readiness to quit | Unclear |
| Reminders | ||
| McClure et al. 2013 | Yes | |
| Muñoz et al. 2009 | Yes | |
| Strecher et al. 2008 | Yes | |
| Danaher et al. 2013c | Unclear | |
| Houston et al. 2015c | Unclear | |
| Severson et al. 2008c | Unclear | |
| Delivery strategies | ||
| Lieberman et al. 2006 | Multimedia (imagesa vs. text) | Yes |
| Stanczyk et al. 2013 | Multimedia (videoa vs. text) | Yes |
| Strecher et al. 2008 | Single exposurea vs. staged | Yes |
| Schulz et al. 2013d | Single exposure vs. alternating feedback | Unclear |
| McClure et al. 2013d | 1. Dictated order of content vs. not 2. Message tone (prescriptive/motivational) | Unclear |
| Social support | ||
| Houston et al. 2015 | Peer + therapist support | No |
| Danaher et al. 2013c | Peer support | Unclear |
| Muñoz et al. 2009 | Peer support | Unclear |
| Schaub et al. 2015d | Therapist support | Unclear |
| Severson et al. 2008c | Peer + therapist support | Unclear |
| Stoddard et al. 2008d | Peer support | Unclear |
| Incentives | ||
| Ramo et al. 2015d | Contingent on website comments | Unclear |
| Stoops et al. 2009 | Contingent on abstinence | No |
a
Independent effects reported for specific sub-types.
b
Statistical tests not conducted.
c
These studies used a multi-component design so eliciting individual effects was not possible.
d
Inconsistent findings reported between engagement measures.