Table 2.

Randomized-controlled trials of interventions to improve engagement and retention: continuous and integrated strategies

Study	Sample/design/setting	Intervention	Measures	Impact of intervention on outcomes?
				Engagement	Retention
Heinrichs (2006)	Families who were offered parenting intervention at 15 preschools in disadvantaged neighborhoods in Germany Preschools matched on size, then randomized at preschool level	Payment and group versus individual family sessions I = payment for each family home-based session and bonus for attending all 8 1-h individual home-based sessions (n = 96) I-2 = payment for each group session and phone contact, and bonus for attending all 4 2-h group sessions (n = 59) I-3 = no payment for individual family home-based sessions (n = 45) I-4 = no payment for group sessions (n = 48)	Engagement # of program hours attended Retention % completed program	No	No
Cunningham et al. (1995)	150 parents of kindergarten children with conduct problems offered treatment in Canadian school district Block randomization (marital status, sex, and severity of aggression)	Group versus individual family sessions I = 12-session neighborhood-based group parent training program (n = 48) C = 12-session individual family clinic-based parent training program as usual (n = 46) C-2 = wait-list control (n = 56)	Engagement % of scheduled sessions attended Retention % families dropped	No	No
Korfmacher et al. (1999)	480 low-income, first-time mothers recruited from prenatal clinics for a 2.5 year home visiting preventive intervention from pregnancy to child age 2 Stratified randomization (race/ethnicity, gestational age, and location)	Type of provider (e.g. training and background experience) I-1 = 2.5 year home visitation program delivered by nurses (n = 236) I-2 = 2.5 year home visitation program delivered by paraprofessionals (n = 244)	Engagement Number of completed visits Number of unsuccessful attempted home visits Retention % families dropped	Yes I-1 > 1–2(28 vs. 23 completed visits) I-1 \ I-2 (5.35 vs. 7.63 attempted visits in infancy)	Yes I-1 > I-2 (38 vs. 48%) (ITT)
Prinz and Miller (1994)	147 families with child conduct problems requesting treatment at outpatient family therapy clinic Stratified randomization (single parent status, SES, age of child)	Augmented parent stress and coping skills intervention I = 24-session child management therapy & added therapist support discussions (n = 72) C = child management therapy as usual (n = 75)	Engagement % completing “some” sessions (undefined) Retention % completed treatment (undefined)	Yes I > C (no % reported)	Yes I > (71 vs. 53%) (ITT)
Kazdin and Whitley (2003)	127 families with children with conduct problems requesting treatment at outpatient child mental health clinic (subgroup of families who completed treatment, n = 84) Consecutive referrals were randomized to conditions	Augmented parent stress and coping skills intervention I = 20–25 sessions of parent and child management treatment and 5 sessions focused on reducing parent stress (n = 70 randomized, unclear n for completer subgroup) C = 20–25 sessions of parent and child management treatment only (n = 57 randomized, unclear n for completer subgroup)	Engagement # of cancellations divided by # of weeks in treatment Retention # of no-shows divided by # of weeks in treatment	No	N/A
Miller and Prinz (2003)	124 families with child conduct problems requesting treatment at University outpatient family therapy clinic Block randomization (SES, age of child)	Augmented parent stress and coping skills intervention in combination with parent and/or child sessions I = 24-session parent training & added parent support discussions (n = 32) I-2 = 24-session child cognitive treatment (n = 31) 1–3 = conditions 1 ? 2 (n = 30) 1–4 = condition 1 ? child relationship support (n = 31)	Engagement % kept appointments % late arrivals Retention Premature termination (parent declined or missed 3 consecutive appts)	Yes I-2, I-3, I-4 > I (% not reported)	Yes I-2, (I-3 ? I-4) \I (19, 31% vs. 50% dropout) (ITT)
Nock and Kazdin (2005)	76 families with child conduct problems requesting treatment at outpatient family therapy clinic Block randomization (blocks unreported)	Engagement MI sessions I = 8-session parent training plus three brief MI engagement-focused discussions across treatment (n = 39) C = 8-session parent management training as usual (n = 37)	Engagement # of completed sessions Reported adherence to treatment Retention Treatment completion Premature termination (\ 8 sessions)	Yes I > C (6.4 vs. 5.2 completed sessions) I > C greater adherence in later sessions (subgroup analyses)	Yes I > C (56 vs. 34%) (ITT) I \ C premature termination
Mullins et al. (2004)	71 low-income women mandated to 12-months outpatient substance use treatment by child welfare due to prenatal drug use at University free mental health clinic Stratified (not reported) randomization	Brief drug-use focused MI sessions I = outpatient group treatment for drug use with 3 MI sessions during early stages of treatment (n = 35) C = outpatient group treatment for drug use, 2 educational video sessions and one home visit (n = 36)	Engagement # of group sessions attended Retention Completion of treatment through the 3 MI sessions (I) or 2 educational sessions plus home visit (C)	No	No
Grote et al. (2007, 2008, 2009); Swarz et al. (2007)	53 low-income, depressed mothers and expectant mothers requesting treatment in an outpatient clinic Block randomization (race)	Multi-component intervention I = MI and ethnographic interviewing engagement session and on-going case management in addition to adapted Interpersonal Therapy (8 sessions plus as- needed booster sessions) (n = 25) C = adapted Interpersonal Therapy (8 sessions plus as- needed booster sessions) (n = 28)	Engagement # of completed sessions Retention % completed 7–8 sessions	Unclear I > C (# not reported)	Yes I > C (68 vs. 7%) (ITT)

Note: N/A not assessed/not applicable, I intervention condition, C control condition, ITT intent-to-treat analyses. We note when there is more than one intervention condition with I-2, I-3, or I-4