Table 4 Retention strategies described in articles reviewed

Davis et al. [15]

Run-in phase

Weekly grocery gift cards

–

Transportation support

–

79.4 % (54/68)

Ventura et al. [16]

Azevedo et al. [17]

–

Rewards for retention

–

–

–

100 % (252/252)

Berry et al. [18–20]

1) Delayed intervention for control group

2) Counseling or support

1) Exercise equipment

2) Money for data collection

3) Food

4) Gifts

1) Regular contact

2) Refrigerator magnet

3) Building staff–participant relationship

1) Reminder message

2) Flexible data collection days

3) Childcare

4) Transportation support

1) Complete contact information

2) Toll-free line

3) Tracking letter

89.1 % (638/716)

Elizondo-Montemayor et al. [21]

–

–

Building staff–participant relationship

Reminder message

–

80.8 % (101/125)

Black et al. [24, 25]

Culturally sensitive

–

–

–

–

78.3 % (184/235)

Hurley et al. [26]

Witherspoon et al. [27]

Weigensberg et al. [28]

–

–

Transportation support Make-up session

–

82.9 % (29/35)

Ritchie et al. [33]

1) Alternative intervention for control group

2) Counseling or support

3) Culturally sensitive

1) Exercise equipment

2) Recipe books

1) Building staff–participant relationship

2) Regular contact

Transportation support

–

57.9 % (136/235)

Sharma et al. [34]

Burnet et al. [37]

1) Culturally sensitive

2) Activities at YMCA and grocery stores

–

Building staff–participant relationship

1) Convenient intervention sites

2) Transportation support

3) Child care

–

62.1 % (18/29)

Davis et al. [38–40]

–

1) Weekly prizes

2) Increasing money for data collections

3) Food at intervention session

Regular contact

Transportation support

–

94.1 % (209/222)

Tkacz et al. [41]

Petty et al. [42]

Wickham et al. [44]

–

YMCA membership

–

–

–

–

Bean et al. [45]

–

1) YMCA membership

2) Grocery store gift card for data collection

–

–

–

–

Wysocki et al. [46, 47]

Alternative intervention for control group

1) Money for each data collection

2) Money for completing all intervention sessions

–

–

–

96.6 % (115/119)

Wysocki et al. [48–50]

Alternative intervention for control group

1) Money for each data collection

2) Money for completion of all intervention sessions

–

–

–

88.5 % (92/104)

Ellis et al. [51, 52]

Alternative intervention for control group

–

–

Convenient intervention sites

–

92.9 % (118/127)

Story et al. [2]

1) Alternative intervention for control group

2) Fun intervention activities

3) Culturally sensitive

1) Gift for intervention attendance

2) Money

3) Increasing money for data collections

4) Additional money for blood draw

5) Food

1) Family nights

2) Regular contact

3) Build relationship between study and broader community

1) Convenient intervention sites

2) Flexible study procedures and measurement visits

3) Home visits for data collection

4) Transportation support

5) Childcare

6) Email and telephone reminders

1) Complete contact information

2) Tracking database

3) Calls from ‘non-identifiable’ cell phones

Phase I:

Rochon et al. [53]

ᅟ

Kumanyika et al. [54, 55]

91.4 % (32/35) and 100 % (60/60)

Klesges et al. [56, 57]

Phase II:

Robinson et al. [58, 59]

80.2 % (243/303) and 86.2 % (225/261)

Stockton et al. [60]

Natale et al. [61]

Alternative intervention for control group

Incentives (not specified)

Regular contact

–

–

–

Nansel et al. [62]

Alternative intervention for control group

1) Money for completing all data collections

2) Additional money for child providing blood glucose meter data

1) Appointment reminder calls

2) Follow-up calls after appointment

1) Transportation support

2) Midpoint evaluations by telephone

–

92.3 % (360/390)

Janicke et al. [63]

1) Delayed intervention for control group

2) Proper participant goals

3) Person-centered intervention

1) Drawing for gift card at weekly child session

2) Gift card per family for each session

3) Money for data collections

4) Food

1) Build community connections

2) Regular contact

3) Phone calls to participants after missed sessions

Make-up sessions

–

87.1 % (81/93)

Follansbee-Junger et al. [64]

Radcliff et al. [65]