Table 1.

Study characteristics.

Author, year and location	Targeted substance/ health behaviour	Study characteristics	Description of DI	n participants	Characteristics of study population	Engagement outcome measure(s)	Reported engagement results
Tailoring
Danaher et al. 2012 USA	Smokeless tobacco	Design: RCT (three arms) Participants: adult smokers Arms: (1) active control (basic version), (2) enhanced, (3) enhanced + EPS(s): tailoring/social support/reminders/multimedia recruitment: referral from smoking clinic DI format: web-based	Name: MyLastDip Arm 1 (basic version): included step-by-step instructions for best practices for cutting down and a resources section Arm 2 (enhanced version): provided above interactively with multimedia, personalised content; email reminders; goal setting and a web blog	N = 1716 (Enhanced = 857; basic = 859)	% male: 97% % White: 96% Age: 20.8 years, SD = 2.6 Readiness to quit: mean of 9 (SD 2.1) on an 11 point scale	(1) Logins (2) Duration of visit (min)	(1) Participants in enhanced version logged in more compared with basic** (2) Participants in enhanced version had higher duration of visit (min) compared with basic***
Elfeddali et al. 2012 Netherlands	Tobacco	Design: RCT (three arms) Participants: adult smokers aged 18–65 years Arms: (1) basic version (action planning (AP) programme), (2) enhanced version (AP+), (3) control (questionnaire only) EPS(s): tailoring (basic vs. enhanced tailoring) Recruitment: online and print DI format: web-based	Name: Stay Quit for You (SQ4U) Tailored smoking cessation web-based programme using the I-Change Model58 Arm 1: basic included (AP): (1) tailored feedback before the quit attempt, (2) tailored planning strategy assignments before and after the quit attempt Arm 2: enhanced included above, plus tailored feedback after the quit attempt Arm 3: questionnaire only	N = 1812 (after exclusions) (Basic tailoring: n = 542, enhanced tailoring: n = 622, control: n = 648)	% male: 38% % White: not reported Age: 41 years (11.80) Readiness to quit: not reported	Number of planning assignments completed (six in total)	Participants in enhanced condition (AP+) completed more planning assignments than participants in the basic condition (AP)*
Houston et al. 2015 USA	Tobacco	Design: RCT (two arms) Participants: adult smokeless tobacco users Arms: three arms of increasing complexity EPS(s): interactivity/tailoring/social support/reminders/multimedia recruitment: online and print paid advertisements, radio DI format: web-based	Name: Decide2quit Arm 1: interactive, tailored quit smoking website. Including: motivational messaging tailored to readiness to quit; interactive risk, decisional balance, and cessation barrier calculators; informational resources Arm 2: enhanced version included above, plus a tailored pushed motivational email messaging system Arm 3: enhanced version + included the above ((1) and (2)) plus a social support group and communication with a trained tobacco treatment specialist	N = 900 (control = 299, enhanced = 164, enhanced+ = 437)	% male: 37% % White: 85% Age: 50% were 35–55 years (mean and SD not reported) Readiness to quit: 80% thinking about quitting	(1) Logins	(1) Participants in enhanced+ logged in more than the control*** (2) Enhanced+ vs. enhanced was non-significant
McClure et al. 2013 USA	Tobacco	Design: randomised fractional factorial design with four two-level intervention factors Participants: adult smokers Factors: participants randomised to one of two levels of each four factors: (1) Message tone (prescriptive vs. motivational), (2) navigation autonomy (dictated vs. not), (3) proactive email reminders (yes vs. no), (4) personally tailored testimonials (yes vs. no) EPS(s): tailoring/reminders recruitment: identified from automated health plan records DI format: web-based	Name: Q2 Tailored (to stage of readiness to quit and demographics) Internet based smoking cessation programme offering material addressing: the risks of smoking; benefits of quitting smoking; perceived barriers to treatment, perceived barriers to quitting; strategies for reducing smoking Factor 1: message tone (prescriptive/didactic vs. motivational) Factor 2: navigation autonomy (dictated order vs. not) Factor 3: proactive email reminders (yes vs. no) Factor 4: inclusion of personally tailored testimonials (yes vs. no)	N = 1865	% male: 37% % White: 82% Age: 44.2 years (SD = 14.7) Readiness to quit: in next 30 days: 44%	(1) Website visits, (2) content pages viewed, (3) minutes online, (4) content areas viewed	(1) Tailored testimonials: no effect (2) Email reminders increased visits***, page views***, min online*** not content area views (3) Prescriptive message tone increased page** and content*** area views but not visits or min online (4) Dictated navigation increased content area*** and page views**, min online***, not visits
Severson et al. 2008 USA	Smokeless tobacco	Design: RCT (two arms) Participants: adult smokeless tobacco users Arms: (1) enhanced version, (2) basic version EPS(s): interactivity/tailoring/social support/reminders/multimedia recruitment: online and print paid advertisements, radio DI format: web-based	Name: Chewfree.com Arm 1 (basic version): included best practices to cut down and cessation information Arm 2 (enhanced version): arm 1 plus guided, interactive and tailored ‘planning to quit module’; ‘staying to quit’ module; multimedia content; two web-based support groups (peer and research staff; email reminders	N = 2523 (Enhanced = 1260; basic = 1263)	% male: 98% % White: 98% Age: 37 years, SD = 9.6 Readiness to quit: reported a mean of 8 (SD 1.8) on an 11 point scale (ref.)	(1) Logins (2) Duration of visit (min)	(1) Participants in enhanced version logged in more compared with basic*** (2) Participants in enhanced version had higher duration of visit (min) compared with basic***
Strecher et al. 2008 USA	Tobacco	Design: Randomised fractional factorial design with five two-level intervention factors Participants: adult smokers Factors: participants randomised to one of two levels of each of the five factors: (1) outcome expectations, (2) efficacy expectations messages, (3) use of hypothetical success stories, (4) personalisation of the message source, (5) timing of message exposure EPS(s) tailoring/feedback strategies: Recruitment: identified from automated health plan records DI format: web-based	Name: not reported Core-intervention based on cognitive-behavioural methods of smoking cessation and relapse prevention including motivations for quitting, stimulus control and self-efficacy enhancement Factor 1: tailored outcome feedback expectations (high/low) Factor 2: tailored efficacy expectations messages (high/low) Factor 3: use of hypothetical success stories (high/low) Factor 4: personalisation of the message source (high/low) Factor 5: timing of message exposure (all at once/over a five week period with reminder)	N = 1866	% male: 40% % White: 79% Age: 46 years (SD = not reported) Motivation to quit (mean on 1–10 scale): = 8.3	Number of web-based smoking cessation sections opened by the participant	Regression analyses revealed that participants receiving high-depth tailored self-efficacy components*** and single exposure information opened significantly more web-sections***
Tensil et al. 2013 Germany	Alcohol	Design: RCT (two arms) Participants: adult drinkers (18 or over who drank at level considered harmful or consumed more than 24/12 g (male/female) of pure alcohol per day on average in the past week) Arms: (1) original version, (2) revised version EPS(s): tailoring (original vs. enhanced) Recruitment: email DI format: web-based	Name: Change Your Drinking Internet-based self-help programme with automated tailored feedback. Based on solution-focused brief intervention59 and cognitive behavioural therapy Arm 1: includes drinking self-assessment, tailored feedback, goal planning, drinking diary and goal feedback based on motivational interviewing techniques60 Arm 2: revised version of above including more intensive daily short tailored feedback on the individual’s alcohol use and graphical display of alcohol use, detailed tailored feedback on alcohol use and tips on how to cope with risk situations on day 7, and advice for reflecting on reasons for reducing drinking and advice on rewarding achievement of goals	N = 595 (Original = 300, revised = 295)	% male: original = 59%, revised = 63% % White: not reported Age: original mean = 29.8 (10.3), revised = 29.0 (9.4) Measure of motivation to cut down: not reported	Diary usage (dichotomous: used at least once)	The usage of the diary was significantly higher*** in the revised version of the programme (diary used at least once, original version n = 167 (55.7%), revised n = 238 (80.7%)
Reminders
Danaher et al. 2012 USA	See Danaher above
Houston et al. 2015 USA	See Houstan above
McClure et al. 2013 USA	See McClure above
Muñoz et al. 2009 USA	Tobacco	Design: RCT (four arms) Participants: adult Spanish/English speaking internet users who smoked cigarettes Arms: (1) basic version, (2) basic + email reminders, (3) basic + email reminders + mood management lessons, (4) basic + email reminders + mood management lessons + social support EPS(s): reminders/social support/online therapist guidance Recruitment: online/word of mouth DI format: web-based	Name: or a Arm 1: static online smoking cessation guide covering: reasons to quit, cessation strategies, relapse prevention and management, pharmacological aids. Also included a cigarette counter and online journal Arm 2: Basic version + automated emails with links to sections of the GUIA keyed to quit date Arm 3: arms 1 + 2 and eight-lesson cognitive-behavioural mood management course Arm 4: arms 1 + 2 + 3+ online support group	N = 1000 (Arm 1 = 247; arm 2 = 251; arm 3 = 251; arm 4 = 251)	% male: 55% % White: 70% Age: 38 years, SD = 11.3 Confidence in quitting (as proxy for readiness to quit): mean = 6.8 on a 10 point scale (SD = 2.0)	As a proxy for engagement (1) cigarette counter usage, (2) online journal usage	Conditions with email reminders (2, 3 and 4) had increased cigarette counter*** and journal usage** compared with condition 1 (data not reported). Conditions 3 and 4 had increased cigarette counter usage compared with condition 2***
Strecher et al. 2008 USA	See Strecher above
Severson et al. 2008 USA	See Severson above
Delivery strategies
Lieberman et al. 2006 USA	Alcohol	Design: RCT (two arms) Participants: adult drinkers Arms: (1) text based feedback, (2) multimedia based feedback EPS(s): feedback strategies (text vs. multimedia) Recruitment: online DI format: web-based	Name: Alcohol Check up Web-based alcohol screening and personalised feedback Arm 1: feedback presented in HTML formatting Arm 2: multimedia feedback that included an animated photograph of a woman’s face which personified the programme as she guided the user through the feedback modules	N = 288 Between arms not reported	% male: text = 62.8%, multimedia = 69% % White: text = 87.2%, multimedia = 86.8% Age: text mean = 37.2 (11.8), multimedia = 36.0 (12.1) Measure of motivation to cut down: not reported	Feedback modules viewed	Participants in the personified guide arm (multimedia) viewed significantly** more viewed feedback modules (3.9) than participants in the text-only group (3.7)
McClure et al. 2013 USA	See McClure above
Schulz et al. 2013 Netherlands/Germany	Alcohol	Design: RCT (three arms) Participants: adult drinkers (18 or over who drank at level considered unhealthy Arms: (1) wait list control, (2) summative feedback, (3) alternating feedback EPS(s): delivery (feedback) strategies Recruitment: online DI format: web-based	Name: Alcohol-Everything Within the Limits?! Three session, tailored programme targeting adult problem drinkers based on the I-Change model58 including personalised feedback and advice for cutting down drinking Arm 1: waiting list control Arm 2: summative feedback. All feedback provided at once after answering questions Arm 3: alternating feedback where participants received questions and personal advice alternately	N = 448 (Control = 135, summative feedback = 181, alternating feedback = 132)	% male: 57% % White: not reported Age: mean = 41.7 years old (SD = 15.7) Measure of motivation to cut down: not reported	(1) Programme completion	After the first session, drop out was significantly less in the alternating condition**. However, no differences in programme completion at three and six months
Stanczyk et al. 2013 Netherlands	Tobacco	Design: RCT (two arms) Participants: smokers, 16 years or older, motivated to quit, categorised as lower or higher educated participants Arms: (1) text based messages, (2) multimedia based messages EPS(s): delivery strategies (text vs. video) Recruitment: online DI format: web-based	Name: Steunbij Stoppen Web-based tailored smoking cessation programme based on the I-Change Model58 Arm 1: (text) tailored messaging delivered by text Arm 2: (video) tailored messaging delivered by video. All content between the two arms was identical	N = 204 (Text: n = 104, video: n = 100)	% male: text = 40.5, video = 22 % White: not reported Age: text mean = 46.6 (11.9), multimedia = 48.2 (12.0) Readiness to quit within one month (%): text = 19.2, video = 18.5	Duration of visit (min)	Text condition spent a mean of 7.2 min online, video condition spent a mean of 7.8 min online***
Strecher et al. 2008 USA	See Strecher above
McClure et al. 2013 USA	See McClure above
Social support
Danaher et al. 2012 USA	See Danaher above
Houston et al. 2015 USA	See Houston above
Muñoz et al. 2009 USA	See Muñoz above
Schaub et al. 2015 Switzerland	Cannabis	Design: RCT (three arms) Participants: adult (18 years and older) using cannabis at least once a week over the 30 days prior to study entry Arms: (1) original (web-based self- help programme), (2) original + chat counselling, (3) wait list control EPS(s): therapist guidance (online chat counselling) Recruitment: online and print DI format: web-based	Name: CanReduce Web-based tailored self help programme based on CBT,61 MI60,62 approaches consisting of eight modules delivered over a six week period. Arm 1: web-based self-help programme with up to two individual chat counselling sessions based on MI and CBT Arm 2: web-based self- help programme without chat counselling Arm 3: wait list control with participants receiving intervention after three months	N = 308 (Original = 114, original + chat = 101, control = 93)	% male: 75.3% % White: not reported Age: mean = 29.8 years old (SD = 10.0) Measure of motivation to cut down: not reported	(1) Module completion (2) Diary consumption completion	(1) Participants in the chat counselling arm (arm 1) did not complete significantly more modules than in the original arm (arm 2) (2) Participants in the chat arm completed significantly* more diary entries than in the original arm
Severson et al. 2008 USA	See Severson above
Stoddard et al. 2008 USA	Tobacco	Design: RCT (two arms) Participants: adult tobacco smokers Arms: (1) basic, (2) basic + social support (BB) EPS(s): peer support Recruitment: email to federal employees DI format: web-based	Name: Smokefree.gov Web-based self-help smoking cessation programme including online quit guide, self-help modules, 1:1 counsellor smoking cessation support, interactive risk tools, evidence-based risk tools and lists of clinical trials still recruiting smokers Arm 1 (basic): participants received the above intervention only Arm 2 (BB): participants received access to a forum where they could interact with other users	N = 1375 (basic = 691; BB = 684)	% male: 46.1 % White: 69.1 Age: 43.6 (SD = 10.3) Readiness to quit: not reported	(1) Duration of visit (min) (2) Visits to pages (tools)	(1) Participants in the BB group spent more time in minutes (mean = 18) on the intervention compared with participants in the basic group (mean = 11) (2) Visits to different pages did not notably differ between the basic and BB conditions, statistical tests were not conducted
Incentives
Ramo et al. 2015 USA	Tobacco	Design: RCT (two levels of randomisation: three arms randomised to Facebook group by stage of change, within each arm randomised to one of three incentive conditions) Participants: young adult smokers (18–25 years old) Facebook group arms: (1) not ready to quit, (2) thinking about quitting, (3) ready to quit Incentive arms: (1) no incentive, (2) personal condition: $50 gift card, (3) altruistic incentive: $50 gift card to charity EPS(s): incentives Recruitment: Facebook advertisement DI format: web-based	Name: Tobacco Status Project Description: a Facebook intervention tailored to stage of change; Transtheoretical Model strategies60 for cessation; group messaging (posts); CBT cessation sessions; ask the doctor sessions Arm 1: no incentive Arm 2 (personal incentive): $50 gift card incentive emailed to participant for commenting on all 90 posts posted to their Facebook group Arm 3 (altruistic incentive): $50 gift card incentive donated to charity for commenting on all 90 posts posted to their Facebook group	N = 79 Arm 1 (not ready to quit): n = 35 randomised to personal condition: 8, altruistic condition: 11, no incentive: 16 Arm 2 (thinking about quitting): n = 32 Personal condition: 9, altruistic condition: 8, no incentive:15 Arm 3 (ready to quit): n = 12 Personal condition: 3, altruistic condition: 3, no incentive: 6	% male: 80% % White: 80% Age: 21 years, SD = 2.1 Smoking goal: 10% reported abstinence goal, 60% reduction goal, 30% no goal	As a proxy for engagement number of ‘likes’ and comments measured	(1) For enhanced sample: no significant difference among incentive conditions on number of comments made to Facebook groups (2) For those who commented at least once: personal incentive condition made more comments than other two conditions*
Stoops et al. 2009 USA	Tobacco	Design: RCT (two arms) Participants: adult tobacco smokers Arms: (1) Abstinence Contingent group (AC), (2) Yoked Control group (YC), EPS(s): Incentives Recruitment: local advertisements, word of mouth DI format: web-based	Name: Not given Intervention was an internet-based CO breath recorder which provided feedback and progress tracking Arm 1 (AC): participants received monetary incentives contingent on recent smoking abstinence (CO level of ≤4 parts per million) Arm 2 (YC): participants received monetary incentives independent of smoking status. Participants were matched to a participant in the AC group and were reinforced on a schedule identical to that of their ‘yoked’ partner	N = 68 (AC = 35; YC = 33)	% male: AC group: 26%; YC group: 24% % White: AC group: 94%; YC group: 94% Age: AC group: mean = 38 years, range = 21–58; YC group: mean = 40, range = 18–61 Readiness to quit: not reported	As a proxy for engagement number of videos posted to website was used as a measure of engagement	No significant effect of group on odds of posting videos to the website. AC group posted 68% of the total number of videos expected, YC group posted 67% of the total number of videos expected

Note: *=<.05, **=<.01, ***=<.001. RCT: randomized controlled trial; EPS: engagement promoting strategy; DI: Digital Intervention; CBT: cognitive behavioural therapy; MI: Motivational Interviewing