. 2020 Jun 3;9:128. doi: 10.1186/s13643-020-01385-8

Table 1.

Characteristics of included studies

Study ID Country Related publications	Study design, duration of the intervention	Population and setting Number of participants (n) Age (years) Female/male	Description of the intervention Intervention group (IG) Control group (CG)	Outcome measures (primary outcome, secondary outcome)	Key findings	Effect size for physical activity (PA) [95% confidence interval]
Ballesteros 2014 [42] Spain, Sweden, Greece	RCT, 12 months	Communities in Spain, Sweden, and Greece n = 41 Age range 65–85 GI: mean age 74, GC mean age 75 Female/male: IG 16/9 CG 11/5	ICT-mediated social network: AGNES IG: AGNES CG: chat and coffee with the research team	Wellbeing (SPF-IL scale)	IG improved affective dimension (p < 0.05) at post-test. IG improved affective dimension 8.92 (SD 1.93) and 10.20 (SD 1.44), at pre- and post-test, respectively	No data on PA
Cook 2015 [24] USA	RCT, 3 months	Workers aged 50 years and older n = 278 Age range 50 to 68 Female/male: IG 40/98, CG 50/89	Web-based multimedia program (information and guidance) IG: web-based multimedia program CG: waitlist	Diet change, mild exercise, self-efficacy Eating practices, exercise planning, beliefs about aging	IG performed better on diet change (p 0.048), planning healthy eating (p 0.03), and mild exercise (p 0.01). IG vs CG showed effects on eating practices (p 0.03), exercise self-efficacy (p 0.03), exercise planning (p 0.03), and aging beliefs (p 0.01)	Not possible to calculate effect size
Homma 2016 [52] Japan	Pilot RCT, 3 months	Two districts of Kurihara city n = 68 Mean age: IG 65.1, GC 67.2 Female/male: IG 22/13, CG 22/11	IG: videophone group (interactive interviews) CG: document group (printed communication) Telemonitoring of health conducted in both groups	Physical activity, behavioral change self-assessment (PA and Diet) Clinical parameters (body weight, BMI, blood pressure, albumin) Perceived health condition and improved lifestyle	Both CG and IG improved average step per day: CG 5046 vs. 5992 (p < 0.01), IG 5829 vs. 7324 (p < 0.01); between group (p = 0.16). IG improved behavioral change for PA (p 0.004), diet (p 0.002), and lifestyles (p < 0.01). IG improved significantly in most clinical parameters such as blood pressure, HbA1c, albumin, BMI. IG perceived higher improvement in health condition and lifestyle (72.7% vs. 97.1% (p < 0.01).	0.21 [− 0.28–0.70]
Irvine 2013 [43] USA	RCT, 12 weeks	Sedentary men and women 55 years and over, community n = 368 Mean age 60.3 (SD 4.9) Female/male: IG 127/51, CG 129/61	Web-based intervention to promote physical activity IG: web-based intervention CG: no access to website	Physical activity Body mass index Quality of life SF-12 health survey	IG improved on 13 of 14 outcome measures. IG maintained large gains on all 14 outcomes measures after 6 months.	0.28 [0.05–0.51]
Kim 2013 [44] USA	RCT, 6 weeks	African-American community n = 46 Mean age: GI 69.3 (SD 7.3), GC 70.5 (SD 7.5) Female/male: IG 21/5, CG 8/2	Text messaging to motivate walking IG: pedometer, walking instructional manual and text messaging CG: without text messaging	Physical activity Step count Perceived activity levels Leisure, time, exercise, questionnaire (LTEQ)	IG improved steps vs. CG (679 vs. 398; p < 0.05), as well as LTEQ score (p < 0.05). Both groups increased their LTEQ score at 6 weeks (p < 0.001).	0.12 [− 0.63–0.88]
Kurti 2013 [45] USA	Quasi experimental (controlled trial), 2 months	Community members over 50 years in Florida n = 12 Mean age 65.5 Female/male: IG 5/1, CG 5/1	Internet-based intervention (successive 5-day blocks) to increase physical activity in sedentary adults IG: monetary consequences CG: no monetary consequences	Physical activity	IG and CG reached the 10,000-step goal. IG vs. CG increased steps (182% vs. 108%) and met steps goals (87% vs. 52%).	Not possible to calculate effect size
Lara 2016 [33] UK	RCT, 8 weeks	Workplaces in Northeast England n = 75 Mean age 61 (SD 4) Female/male: IG 38/12, CG 19/6	Web-based intervention (LEAP) IG: LEAP CG: use NHS choices website, UK Department of Health	Physical activity, Mediterranean diet (MD adherence)	Both IG and CG improved outcomes and no significant differences were detected.	Not possible to calculate effect size
Mouton 2015 [46] Belgium	RCT, 4 arms	One municipality in Belgium n = 204 Mean age 65 Female/male: IG1 20/13, IG2 27/13, IG3 25/13, CG 23/15	Web-based, center-based or combined physical activity (PA) intervention IG1: web-based intervention IG2: center-based intervention IG3: mixed (center- and web-based) intervention CG: no intervention	Physical activity (PA) level Readiness for PA, awareness of PA (general), awareness of PA (opportunities in municipality)	IG3 improved in PA level (p 0.041), readiness for PA (p 0.001). IG3 improved on awareness of PA (p 0.003) and awareness of PA opportunities in municipality (p 0.001).	0.06 No data available to calculate CI
Myhre 2013 USA	RCT, 3 arms, 8 weeks	2 cohorts from retirement communities in Arizona n = 41 Mean age 79.4 Female/male: IG1 9/5, IG2 9/4, CG 11/3	Micro-blogging shared with others or kept private IG1: Facebook IG2: online diary CG: waitlist	Knowledge, letter memory, keep track	IG1: knowledge, memory task improved at time 2 vs. baseline (p < 0.01); keep track slightly improved (p < 0.10)	No data on PA
Nyman 2009 [47] UK	RCT, no duration specified	Community in Southampton n = 302 Mean age 70.41 (SD 7.07) Female/male 187/115	Website with tailored advice to undertake strength and balance training (SBT) IG: website with tailored advice CG: generic website	Attitudes to falls-related intervention scale (AFRIS)	No significant differences in attitudes toward SBT. IG participants indicated that advice was relevant (p 0.017) and activities good (p 0.047).	No data on PA
Peels 2013a [48] Netherlands Related publications: Golsteijn 2014 [55], Peels 2012 [56], Peels 2013b [57], Peels 2014a [58], 2014b [59]	Cluster-RCT, 5 arms, 1 year	Community members n = 1729 Mean age 62 Female/male: IG1 127/51, IG2 144/112, IG3 111/113, IG4 93/100, CG 158/152	Printed or web-based tailored physical activity intervention IG1: printed basic IG2: print-delivered with environmental information IG3: web-based basic IG4: web-based with environmental information CG: No advice	Process outcomes (appreciation, understanding of information)	IG1-IG2: printed intervention vs web-based intervention was significantly higher 92.7–98.2% read, 70.1–76.5% kept, and 39.9–56.8% discussed, and better appreciated (6.06–6.91 vs 5.05–6.11 on a scale of 1–10)	0.10 [− 0.04–0.24]
Slegers 2008 [49] Netherland	Feasibility RCT, 4 arms, 12 months	Community in Maastricht n = 236 Age range 64–75 Female/male: ?	Computer training and Internet usage IG 1: training and intervention IG 2: training, no intervention CG1: no training, no intervention CG2: not interested (passive control)	Physical and psychological well-being (SF-36) Social well-being and social network	Most outcomes were not significant. IG participants spent more time on learning new things.	0.24 [− 0.14–0.63]
van het Reve 2014 [50] Switzerland Related publications: Silveira 2013 [60, 61]	Preclinical exploratory trial, 12 weeks	2 institutions for older people and 1 organization providing home nursing care for seniors n = 44 Mean age (years) 75 (SD 6) Female/male: IG1 8/5, IG2 10/4, CG 10/7	A tablet with ActiveLifestyle IG1: social group with tablet IG2: individual group with tablet GC: brochure group	Gait performance (dual-task walking) Physical performance Short physical performance battery (SPPB) Fall efficacy Fall efficacy scale international (FES-I)	IG1 and IG2 improved significantly in single and dual task walking. IG1, IG2, GC showed SPPB improvement (p 0.02) between pre- and post-test. Group difference for FES-I between GC and IG1et IG2 (p 0.04).	No data on PA
Wijsman 2013 [55] Netherlands Related publications: Vroege 2014 [62], Broekhuizen 2016 [63]	RCT, 3 months	Community in Leiden n = 235 Age range 60–70 Mean age: GI 64.7 (SD 3.0), CG 64.9 (SD 2.8) Female/male: IG 47/72, CG 49/67	Internet-based physical activity intervention: Philips DirectLife IG: Philips DirectLife CG: no intervention	Physical activity Moderate-to-vigorous physical activity (MVPA) Metabolic parameters Quality of life (RAND-36)	IG improved PA, weight, waist circumference, insulin and HbA1c (p < 0.001), and MVPA (p < 0.001). IG improved emotional and mental health (p < 0.03) and health change (p < 0.01).	0.58 [0.31–0.85]

Study ID
Country
Related publications

Study design, duration of the intervention

Population and setting
Number of participants (n)
Age (years)
Female/male

Description of the intervention
Intervention group (IG)
Control group (CG)

Outcome measures (primary outcome, secondary outcome)

Key findings

Effect size for physical activity (PA) [95% confidence interval]

Ballesteros 2014 [42]

Spain, Sweden, Greece

RCT, 12 months

Communities in Spain, Sweden, and Greece

n = 41

Age range 65–85

GI: mean age 74, GC mean age 75

Female/male: IG 16/9

CG 11/5

ICT-mediated social network: AGNES

IG: AGNES

CG: chat and coffee with the research team

Wellbeing (SPF-IL scale)

IG improved affective dimension (p < 0.05) at post-test.

IG improved affective dimension 8.92 (SD 1.93) and 10.20 (SD 1.44), at pre- and post-test, respectively

No data on PA

Cook 2015 [24]

USA

RCT, 3 months

Workers aged 50 years and older

n = 278

Age range 50 to 68

Female/male: IG 40/98, CG 50/89

Web-based multimedia program (information and guidance)

IG: web-based multimedia program

CG: waitlist

Diet change, mild exercise, self-efficacy

Eating practices, exercise planning, beliefs about aging

IG performed better on diet change (p 0.048), planning healthy eating (p 0.03), and mild exercise (p 0.01).

IG vs CG showed effects on eating practices (p 0.03), exercise self-efficacy (p 0.03), exercise planning (p 0.03), and aging beliefs (p 0.01)

Not possible to calculate effect size

Homma 2016 [52]

Japan

Pilot RCT, 3 months

Two districts of Kurihara city

n = 68

Mean age: IG 65.1, GC 67.2

Female/male: IG 22/13, CG 22/11

IG: videophone group (interactive interviews)

CG: document group (printed communication)

Telemonitoring of health conducted in both groups

Physical activity, behavioral change self-assessment (PA and Diet)

Clinical parameters (body weight, BMI, blood pressure, albumin)

Perceived health condition and improved lifestyle

Both CG and IG improved average step per day: CG 5046 vs. 5992 (p < 0.01), IG 5829 vs. 7324 (p < 0.01);

between group (p = 0.16).

IG improved behavioral change for PA (p 0.004), diet (p 0.002), and lifestyles (p < 0.01).

IG improved significantly in most clinical parameters such as blood pressure, HbA1c, albumin, BMI.

IG perceived higher improvement in health condition and lifestyle (72.7% vs. 97.1% (p < 0.01).

0.21 [− 0.28–0.70]

Irvine 2013 [43]

USA

RCT, 12 weeks

Sedentary men and women 55 years and over, community

n = 368

Mean age 60.3 (SD 4.9)

Female/male: IG 127/51, CG 129/61

Web-based intervention to promote physical activity

IG: web-based intervention

CG: no access to website

Physical activity

Body mass index

Quality of life

SF-12 health survey

IG improved on 13 of 14 outcome measures.

IG maintained large gains on all 14 outcomes measures after 6 months.

0.28 [0.05–0.51]

Kim 2013 [44]

USA

RCT, 6 weeks

African-American community

n = 46

Mean age: GI 69.3 (SD 7.3), GC 70.5 (SD 7.5)

Female/male: IG 21/5, CG 8/2

Text messaging to motivate walking

IG: pedometer, walking instructional manual and text messaging

CG: without text messaging

Physical activity

Step count

Perceived activity levels

Leisure, time, exercise, questionnaire (LTEQ)

IG improved steps vs. CG (679 vs. 398; p < 0.05), as well as LTEQ score (p < 0.05).

Both groups increased their LTEQ score at 6 weeks (p < 0.001).

0.12 [− 0.63–0.88]

Kurti 2013 [45]

USA

Quasi experimental (controlled trial), 2 months

Community members over 50 years in Florida

n = 12

Mean age 65.5

Female/male: IG 5/1, CG 5/1

Internet-based intervention (successive 5-day blocks) to increase physical activity in sedentary adults

IG: monetary consequences

CG: no monetary consequences

Physical activity

IG and CG reached the 10,000-step goal.

IG vs. CG increased steps (182% vs. 108%) and met steps goals (87% vs. 52%).

Not possible to calculate effect size

Lara 2016 [33]

RCT, 8 weeks

Workplaces in Northeast England

n = 75

Mean age 61 (SD 4)

Female/male: IG 38/12, CG 19/6

Web-based

intervention (LEAP)

IG: LEAP

CG: use NHS choices website, UK Department of Health

Physical activity,

Mediterranean diet (MD adherence)

Both IG and CG improved outcomes and no significant differences were detected.

Not possible to calculate effect size

Mouton 2015 [46]

Belgium

RCT, 4 arms

One municipality in Belgium

n = 204

Mean age 65

Female/male: IG1 20/13, IG2 27/13, IG3 25/13, CG 23/15

Web-based, center-based or combined physical activity (PA) intervention

IG1: web-based intervention

IG2: center-based intervention

IG3: mixed (center- and web-based) intervention

CG: no intervention

Physical activity (PA) level

Readiness for PA, awareness of PA (general), awareness of PA (opportunities in municipality)

IG3 improved in PA level (p 0.041), readiness for PA (p 0.001).

IG3 improved on awareness of PA (p 0.003) and awareness of PA opportunities in municipality (p 0.001).

0.06

No data available to calculate CI

Myhre 2013

USA

RCT, 3 arms, 8 weeks

2 cohorts from retirement communities in Arizona

n = 41

Mean age 79.4

Female/male: IG1 9/5, IG2 9/4, CG 11/3

Micro-blogging shared with others or kept private

IG1: Facebook

IG2: online diary

CG: waitlist

Knowledge, letter memory, keep track

IG1: knowledge, memory task improved at time 2 vs. baseline (p < 0.01); keep track slightly improved (p < 0.10)

No data on PA

Nyman 2009 [47]

RCT, no duration specified

Community in Southampton

n = 302

Mean age 70.41 (SD 7.07)

Female/male 187/115

Website with tailored advice to undertake strength and balance training (SBT)

IG: website with tailored advice

CG: generic website

Attitudes to falls-related intervention scale (AFRIS)

No significant differences in attitudes toward SBT.

IG participants indicated that advice was relevant (p 0.017) and activities good (p 0.047).

No data on PA

Peels 2013a [48]

Netherlands

Related publications: Golsteijn 2014 [55], Peels 2012 [56], Peels 2013b [57], Peels 2014a [58], 2014b [59]

Cluster-RCT, 5 arms, 1 year

Community members

n = 1729

Mean age 62

Female/male: IG1 127/51, IG2 144/112, IG3 111/113, IG4 93/100, CG 158/152

Printed or web-based tailored physical activity intervention

IG1: printed basic

IG2: print-delivered with environmental information

IG3: web-based basic

IG4: web-based with environmental information

CG: No advice

Process outcomes (appreciation, understanding of information)

IG1-IG2: printed intervention vs web-based intervention was significantly higher 92.7–98.2% read, 70.1–76.5% kept, and 39.9–56.8% discussed, and better appreciated (6.06–6.91 vs 5.05–6.11 on a scale of 1–10)

0.10 [− 0.04–0.24]

Slegers 2008 [49]

Netherland

Feasibility RCT, 4 arms, 12 months

Community in Maastricht

n = 236

Age range 64–75

Female/male: ?

Computer training and Internet usage

IG 1: training and intervention

IG 2: training, no intervention

CG1: no training, no intervention

CG2: not interested (passive control)

Physical and psychological well-being (SF-36)

Social well-being and social network

Most outcomes were not significant.

IG participants spent more time on learning new things.

0.24 [− 0.14–0.63]

van het Reve 2014 [50]

Switzerland

Related publications: Silveira 2013 [60, 61]

Preclinical exploratory trial, 12 weeks

2 institutions for older people and 1 organization providing home nursing care for seniors

n = 44

Mean age (years) 75 (SD 6)

Female/male: IG1 8/5, IG2 10/4, CG 10/7

A tablet with ActiveLifestyle

IG1: social group with tablet

IG2: individual group with tablet

GC: brochure group

Gait performance (dual-task walking)

Physical performance

Short physical performance battery (SPPB)

Fall efficacy

Fall efficacy scale international (FES-I)

IG1 and IG2 improved significantly in single and dual task walking.

IG1, IG2, GC showed SPPB improvement (p 0.02) between pre- and post-test.

Group difference for FES-I between GC and IG1et IG2 (p 0.04).

No data on PA

Wijsman 2013 [55]

Netherlands

Related publications: Vroege 2014 [62], Broekhuizen 2016 [63]

RCT, 3 months

Community in Leiden

n = 235

Age range 60–70

Mean age: GI 64.7 (SD 3.0), CG 64.9 (SD 2.8)

Female/male: IG 47/72, CG 49/67

Internet-based physical activity intervention: Philips DirectLife

IG: Philips DirectLife

CG: no intervention

Physical activity

Moderate-to-vigorous physical activity (MVPA)

Metabolic parameters

Quality of life (RAND-36)

IG improved PA, weight, waist circumference, insulin and HbA1c (p < 0.001), and MVPA (p < 0.001).

IG improved emotional and mental health (p < 0.03) and health change (p < 0.01).

0.58 [0.31–0.85]

CG control group, IG1 intervention group 1, IG2 intervention group 2, IG3 intervention group 3, IG4 intervention group 4