Table 8.

Determinants measured in included studies

Author	Determinantsa
Ajrouch et al. [18]	Social network (size, proportion of family, age, proximity, frequency), education, age, gender, race, health limitation and depression
Bartels et al. [19]	Ratio government expenditure / GDP, interest in politics, children, education, marital status, income, liking the neighbourhood
Bekkers [20]	Trust
Broese van Groenou & Van Tilburg [21]	Age, gender, cohort, education, employment status, health status, marital status, size of personal network and church attendance
Choi & Chou [22]	Education, income, health, work status, religion, generative qualities, number of meetings attended, age, marital status, ethnicity and gender
Cramm & Nieboer [23]	Age, gender, ethnicity, marital status, education, social capital, social functioning, cognitive functioning and physical functioning, volunteering at baseline
Curl et al. [24]	Driving status, waves since driving cessation, gender, age, ethnicity, education, marital status, household income, depressive symptoms, chronic conditions, self-rated health, IADL limitations and cognitive ability
Curl et al. [25]	For both the individual and the spouse: driving status, waves since driving cessation, age, ethnicity, education, couple income, cognitive ability, chronic conditions, IADL limitations and self-rated health
Einolf & Philbrick [26]	Marriage (covariates taken into account but no effect size provided are: volunteering at baseline, ethnicity, education, age, health, hours worked, religious attendance, housework hours and children
Hank & Erlinghagen [27]	Gender, age, education, partnership status, employment status, self-rated health, country
Johnston [28]	Religious importance, religious attendance, family income, functional health, employment status, child currently at home, marital status
Lim & Mac Gregor [29]	Age, gender, ethnicity, income, education, marriage, children, social involvement index, voluntary group involvement, religious tradition, religious index, region, volunteering at baseline, number of close friends, ethnicity of friends, religiosity of friends
McNamara & Gonzales [30]	Age, ethnicity, gender, volunteering at baseline, assets, education, income, health, marital status, volunteer status of spouse, like to spend time with spouse, spousal caregiving, parental caregiving, children, employment status, provision of informal help in community, religious attendance
Mike et al. [31]	Age, gender, education, personality traits: conscientiousness, extraversion, agreeableness, neuroticism and openness, current work status
Nesbit [32]	Volunteering at baseline, gender, age, ethnicity, education, birth of child, divorce, death in family
Okun et al. [33]	Volunteer satisfaction and enjoyment, age, gender, race, hours worked p/wk., education, functional limitations, social interaction, attending clubs /organizations and church attendance
Parkinson [34]	Area of residence, country of birth, English proficiency, education, health care insurance, living arrangements, transport, SF36, DSSI, number of visits to healthcare professionals
Pavlova & Silbereisen [35]	Coping strategies for occupational uncertainty, region, community size, gender, education, income, employment status, partnership status and general health
Pavlova & Silbereisen [36]	Perceived activation demands, volunteering at baseline, age and self-rated health
Son & Wilson [37]	Generativity,religious identification, church attendance, spirituality, religious coping, parental religion, parental sociability, education, age, gender, ethnicity, marital status and income
Son & Wilson [38]	Altruistic obligation, civic obligation, religious identification, spirituality, religious coping, public religiosity, parental religion, education, age, gender, ethnicity, marital status, income, employment, physical health, religious tradition, contact frequency with friends
Son & Wilson [39]	Hedonic well-being, eudemonic well-being, social well-being, age, gender, ethnicity, marital status, education, income, employment, church attendance, physical health
Voorpostel & Coffé [40]	Transitions in partnership, transitions in parental status, age, (change in) education, (change in) employment status, volunteering at baseline
Voorpostel & Coffé [41]	Parental separation, parental levels of voting and volunteering, parental occupation and education, young adults living situation, age, gender, schooling, education, occupation and church visits

^aThe determinants listed here are only those determinants for which the association with the outcome is measured longitudinally and are therefore eligible for inclusion in the meta-analysis