Table 1.

Characteristics of each eligible study

Author (year)	Country	Population (N)	Intervention (N)	Control (N)	Follow-Up	Outcome	Main results	Study design and quality
DOMAIN 2-Increasing job retention
Wallen & Mulloy (2006) [29]	USA	50 Factory workers at a medium sized electronics manufacturing plant	Computer-based respiratory safety trainings to young (<44 yrs) and older (44+ yrs; mean age 51 yrs) workers	Comparison of 3 different programs: TXT, TAP and NAP	–	Learning evaluation	Older participants who received instructions with NAP reached significantly higher level of learning than TXT/TAP participants; older learners may benefit from this approach	Cross Sectional Weak
McDonald et al. (2010) [30]	Australia	27 Hospital Nurses	Aged nurses acting as mentors (12 nurses aged between 40 and 70) and young mentees (15)	–	6 months	Ideas and perceptions of mentors collected by interviews	Mentoring programmes, in particular including retirees, can stimulate professional development, personal growth and benefits both in mentors both in mentees.	Collective case study NA
DOMAIN 3-Improving productivity and workability
Karazman et al. (2000) [31]	Germany	122 Tram, bus and subway drivers of the Munich Transportation Authority 45 and older	20 health days training (physical exercise, professional skills training and self-experience)	–	1 year	WAI (stratified according another specific tool-effect typology questionnaire)	Non-significant increase of WAI (except in a subgroup of older participants)	Pre-post Weak
Mackey et al. (2011) [36]	Australia	Academic and administrative employees between 45 and 70 (mean age = 54 yrs) without a physically active lifestyle	12 weeks long flexible, individually targeted walking intervention facilitated by an occupational physiotherapist (32)	Usual activity (32)	12 weeks	Measures for work ability	This worksite based intervention using behaviour change principles produced significant improvements in work ability	RCT (in press) NA
Rutanen et al. (2014) [40]	Finland	123 occupationally active symptomatic menopausal women aged 44–62 (mean age 54)	6 months aerobic exercise intervention 4 times a week, 50 minutes per session, with a progressive increase in intensity (60)	Without intervention (63)	24 weeks	WAI and Questionnaires on the daily physical and mental work strain	Women in the intervention group have lower work strain, but a tendency towards higher WAI.	RCT; Moderate
Koolhaas et al. (2015) [32]	Netherlands	Nurse and administrative personnel (workers and supervisors) from a University and a University Medical Center (mean age 52)	“Staying healthy at work” problem-solving based intervention to achieve improvement, life-long learning or to tackle problems for a sustainable working life. (64)	Usual activity (61)	1 year	WAI and Productivity (the Quality and Quantity method)	None or negative effects were found on main outcomes (respectively on productivity and WAI). However, effectiveness was shown on three of the secondary outcome measures (work attitude, self-efficacy and skill discretion)	NCT; Moderate
De Boer et al. (2004) [33]	Netherlands	116 employees older than 50 years (mean age 53 yrs) at risk for early retirement of a large international company, which develops and manufactures electronic equipment.	Construction of a detailed action plan, consultation of the employee’s supervisors and personnel managers, and, if appropriate, referral to the general practitioner, a medical specialist, or psychologist. (61)	Received care as usual (they were not invited for a consultation but they could always consult their occupational physician on request) (55)	2 years	Work Ability Index, the Utrecht Burn Out Scale, and the Nottingham Health Profile measuring quality of life	After 2 years no significant differences (except for burnout) in work ability, quality of life and early retirement in the intervention group. A significant improvement was found in the mid term analysis.	RCT; Strong
Palumbo et al. (2012) [37]	USA	14 Nurses older than 49 years (mean age 54 yrs) at one academic medical center	On-site Tai Chi classes once a week and practice on their own for 10 minutes each day for 15 weeks. (7, but 6 included in the analysis)	No intervention (7, but 5 included in the analysis)	15 weeks	Productivity	Tai chi group showed a significant improvement in work productivity (+3 %) and seems cost savings (preliminary cost analysis)	RCT; Moderate
Strijk et al. (2013) [39]	Netherlands	Older workers (45 years or older-mean age 52 yrs) from two academic hospitals	The Vital@Work group (367 workers) received a 6-month Vitality Exercise Program, Personal Vitality Coach visits, and free fruit.	No intervention written except information about a healthy lifestyle in general (363 workers)	1 year	Work engagement, productivity and sick leave	No significant differences regarding work engagement, productivity and sick leave were observed	RCT; Moderate
van Dongen et al. (2013) [38]	Netherlands	See-Strijk et al. (2013) [39]-Domain 3			1 year	Costs related to the Vital@Work intervention, Health care utilization, sport, absenteeism and presenteeism	The program was neither cost-effective (COI) nor cost-saving (ROI).	A COI and ROI analysis of Strijk et al. (2013) [39] RCT: NA
Siukola et al. (2011) [14]	Finland	Blue-collar workers aged 55 years or older (mean age 57 yrs) from Finnish food company	A senior programme looking at the specific needs of older worker with work-related arrangements and dispensations (opportunities to alter the content of work, need for rehabilitation or education) (129)	No intervention (229)	6 years	Total sickness absence days and spells of 1–3, 4–7, 8–21 and >21 days	Sickness absence days increased significantly from baseline in both groups. Intervention group had higher risk for short-time sickness absence, with a reduced risk of long-lasting one	NCT; Weak
Goine et al. (2004) [34]	Sweden	Two paper and pulp manufacturing plants. No specific intervention for older worker, but analysis were stratified for 50–59 and 60–64 age classes	PLANT A (1200) implemented an extensive programme of managerial training and vocational rehabilitation activities. It received about four times more financial support than plant B	PLANT B (1600) Without implementing programmes and with less financial support	10 years (1989–98)	Sick leave and disability pensions.	For employees in the upper age groups, relative risk for long-term and very long-term sick leave was and remained elevated after the intervention. The RR of short-term sick leave (1–14 days), was (not significantly) lower in these groups than among the younger employees.	Cohort; Moderate
Härmä et al. (2006) [35]	Finland	Line maintenance unit of a large airline company	Implementation of a very rapidly forward rotating workplace shift system among young (24–44 years) and elderly (45–61 years) maintenance workers	Without intervention	2 years (1.5 years before and 6 months after a new shift system	Sleep wakefulness, well-being and social life of young and older shift workers	The intervention had positive effects on the sleep, alertness and well-being (including social and family life and hobbies), especially for the older shift workers	NCT; weak
DOMAIN 4-Workplace interventions for health promotion and well-being
Strijk et al. (2012) [43]	Netherlands	Older workers (45 years or older-mean age 52 yrs) from two academic hospitals	The Vital@Work group (367 workers) received a 6-month Vitality Exercise Program, Personal Vitality Coach visits, and free fruit.	No intervention written except information about a healthy lifestyle in general (363 workers)	6 months	Lifestyle behaviours (sports, vigorous physical activities and fruit intake) and vitality-related outcomes (aerobic capacity, mental health and the need for recovery after a work day)	The intervention favourably affected the weekly sports activities, the fruit intake and the need for recovery No effects were observed for other outcomes.	RCT; Moderate
Strijk et al. (2013) [39]	Netherlands	See-Strijk et al. (2013) [39]-Domain 3			1 year	The primary outcome was Vitality (the RAND-36 vitality scale for general vitality, and UWES for work-related vitality)	No intervention effects were observed for vitality, even if high yoga compliers significantly increased their work-related and general vitality.	RCT; Moderate
Palumbo et al. (2012) [37]	USA	See-Palumbo et al. (2012) [37]-Domain 3			15 weeks	Several measures for physical and mental health, work-related stress	Tai chi group showed a significant improvement in physical functions and seems (preliminary cost analysis) cost savings	RCT; Moderate
van Dongen et al. (2013) [38]	Netherlands	See-Strijk et al. (2013) [39]-Domain 3			See-van Dongen et al. (2013) [38]-Domain 3
Chen et al. (2014) [44]	Taiwan	108 Workers aged 50+ years (mean age 55 yrs) from small-and medium scale enterprises	Phase I (4 weeks): organizing action groups, individualized planning of behavioral changes, and updating workers’ health knowledge; Phase II (follow-up 20 weeks) emphasized carrying out the planned lifestyle improvements to reduce the risk of metabolic disorders (58)	Without intervention (50)	24 weeks	Major outcomes were changes in lifestyle, anthropometric and blood biochemical variables	The intervention had a significant positive effect on waist circumference, body weight, BMI, physical activity, triglycerides ad HDL-C. However, the intervention did not improve blood pressure, or serum lipid or HbA1c levels, vegetable consumption, time use, or sleep duration, nor the proportions of subjects having metabolic disorders. The control group had a significant time-related decrease in total cholesterol and HDL-C	NCT; Strong
Merrill et al. (2011) [42]	USA	440 young (18–49 yrs) and old (50+ yrs) workers in a small company. Stratification according to age class allow specific analysis for 64 older worker	All employees receive a four-level wellness programs and quarterly screenings, with prizes and incentives for participants.	No Control	3 years (2007–2009)	Selected Health Indicators: blood pressure, flexibility, body fat, body weight	Overall positive effects. Older employees, who had the highest blood pressure and weight at baseline, showed the greatest decreases in blood pressure and weight.	Cohort; Moderate
Mackey et al. (2011) [36]	Australia	See Mackey et al. (2011) [36]-Domain 3				Measures for step count, % body fat, waist circumference, blood pressure, physical activity & psychological wellbeing	This worksite based intervention using behaviour change principles produced significant improvements in physical activity and health status	RCT (in press) NA
Hughes et al. (2011) [41]	USA	423 participants (older support and academic staff at the University of Illinois at Chicago) aged 40 years and older (mean age 51, range 40 to 68) were categorized into 3 study arms	The COACH (150 workers received a Web-based risk assessments with personal coaching support); the RealAge (135 workers received only Web-based risk assessment and behaviour-specific modules)	The control group received printed health-promotion materials (138 workers)	1 year	Dietary behaviours; Physical activity; Stress. Smoking cessation; Body mass index, waist circumference, and weight.	In the COACH group significant amelioration in fruits and vegetables consumption, percentage of energy derived from fat and physical activity. RealAge participants experienced a significant decrease in waist circumference COACH group participants were almost twice as likely to use their intervention as RealAge participants used theirs.	RCT; Moderate
Cook et al. (2015) [45]	USA	50 years of age and older (range 50–68 yrs) employees located in multiple US offices of a global information technology company (278)	HealthyPast50 workers received a Web-based multimedia program containing information and guidance on the major health promotion topics (138)	wait-list control condition (140)	3 months	Measures of healthy aging, diet, physical activity, stress management, and tobacco use	The HealthyPast50 group performed significantly better than the control group on diet behavioural change self-efficacy, planning healthy eating, and mild exercise. There were not significant improvements on eating practices, moderate exercise, and overall exercise.	RCT; Weak
Koolhaas et al. (2015) [32]	Netherlands	See Koolhaas et al. (2015) [32]-Domain 3				Vitality (the single-item vitality scale of the 12-Item Short Form Health Survey)	Negative effects were found on Vitality. Workers in the intervention group had a 0.10 times higher odds of being in a higher vitality category than the persons in the business as usual group.	NCT; Moderate

Abbreviations: NA Not Applicable; TXT text only, TAP text with pictures, NAP text with pictures, and audio narration, RCT Randomized Controlled Trial, NCT Non-Randomized Controlled Trial, WAI work ability index, COI Cost of Illness, ROI Return on Investment