How do grandparents influence child health and development? A systematic review

Social Science & Medicine 239 (2019) 112476 Contents lists available at ScienceDirect Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed Review article How do grandparents influence child health and development? A systematic review T Aalyia F.A. Sadruddina, Liliana A. Pongutaa, Anna L. Zondermanb, Kyle S. Wileya, Alyssa Grimshawa, Catherine Panter-Bricka,∗ a b Yale University, USA The University of Chicago, USA A R TICL E INFO A BSTR A CT Keywords: Generation Family Parenting Child Health Development Research design Policy Grandparents are often a key source of care provision for their grandchildren, yet they are sidelined in caregiving research and policy decisions. We conducted a global, systematic review of the literature to examine the scope and quality of studies to date (PROSPERO database CRD42019133894). We screened 12,699 abstracts across 7 databases, and identified 206 studies that examined how grandparents influence child health and development. Indicators of grandparent involvement were contact, caregiving behaviors, and financial support. Our review focused on two research questions: how do grandparents influence child health and development outcomes, and what range of child outcomes is reported globally? We examined study design, sample characteristics, key findings, and outcomes pertaining to grandchildren’s physical health, socio-emotional and behavioral health, and cognitive and educational development. Our search captured studies featuring grandparent custodial care (n = 35), multigenerational care (n = 154), and both types of care (n = 17). We found substantial heterogeneity in the data provided on co-residence, caregiving roles, resources invested, outcomes, and mechanisms through which “grandparent effects” are manifested. We identified two important issues, related to operationalizing indicators of grandparent involvement and conceptualizing potential mechanisms, leading to gaps in the evidence base. Currently, our understanding of the pathways through which grandparents exert their influence is constrained by limited data on what grandparents actually do and insufficient attention given to interpersonal and structural contexts. We present a conceptual framework to explicitly measure and theorize pathways of care, with a view to inform research design and policy implementation. We underscore the need for more robust data on three indicators of caregiver involvement—contact, behavior, and support—and for careful description of structural and interpersonal contexts in caregiving research. 1. Introduction Childhood experiences—particularly those connected to the caregiving environment—are critical determinants of wellbeing across the life course (Britto et al., 2017; Pérez-Escamilla et al., 2018). Research in this field, however, primarily focuses on parents. Specifically, studies have demonstrated the centrality of mother-child relationships and father engagement (Panter-Brick et al., 2014) in terms of their impact on children's physical, social, emotional, behavioral, and educational outcomes throughout the life course. Over the past two decades, the roles that grandparents play in caring for grandchildren, and the impacts they have on child health and development, have garnered more systematic attention, especially in the U.S. (e.g., Goodman, 2012; Dunifon et al., 2018) and in Asia (e.g., Xu, 2019). Nevertheless, ∗ grandparents are often sidelined in research and policy initiatives, despite the diverse range of primary and secondary caregiving roles they often assume. Programmatic approaches to child health and development have recently called for a paradigm shift in global policy, arguing for “parents, caregivers, and families to be supported in providing nurturing care and protection in order for young children to achieve their developmental potential” (Britto et al., 2017, p. 91). Although grandparental care is not a new social phenomenon, extended lifespans and transformations in family arrangements, such as decreased family sizes and increased maternal employment, have accentuated grandparents' roles as caregivers globally (Bol and Kalmijn, 2016; Dunifon et al., 2018; Mehta and Thang, 2012; Schatz and Seeley, 2015). In the U.S., for example, census data for 2018 show that 7.82% of U.S. children (0-to-18 years) lived with both a parent and Corresponding author. E-mail address: Catherine.panter-brick@yale.edu (C. Panter-Brick). https://doi.org/10.1016/j.socscimed.2019.112476 Received 27 November 2018; Received in revised form 31 July 2019; Accepted 1 August 2019 Available online 07 August 2019 0277-9536/ © 2019 Elsevier Ltd. All rights reserved. Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. grandparent in three-generation families, while 2.31% lived with a grandparent and no parent in skipped-generation families (U.S. Census Bureau, 2018). Similarly, the Pew Research Center (2013) reported that, of the 7.7 million children living with a grandparent, 3 million had a grandparent as their primary caregiver. Notably, the U.S. census question asking whether co-resident grandparents are the caregivers of their grandchildren was only introduced in 2000 (Goodman, 2012). The nature of residential arrangements shows significant variation across ethnic and socioeconomic groups (Dunifon and Kowaleski-Jones, 2007; Dunifon et al., 2014); indeed, research on the political economy of aging has noted that grandparental caregiving is never equal by race, class, age, or gender (Baker et al., 2008). Grandparents are often the first to assume caregiving responsibilities for children in response to parental incarceration, drug use, unemployment, and death among lowincome minority populations, such as African Americans and immigrant communities in the U.S. (Fuller-Thomson and Minkler, 2001; Hayslip and Smith, 2012; Minkler and Fuller-Thomson, 2005). Higher divorce rates, patterns of economic stagnation, opioid use, and expansion of single-parent homes are also increasing reliance on grandparent caregivers among middle-class white Americans (Gorman and Braveman, 2008; Smith, 2018). Similarly, in Europe, higher numbers of women entering the formal workforce and prohibitive childcare costs have resulted in an increase in grandparents who act as primary or partial caregivers to grandchildren (Eli et al., 2016; Hank and Buber, 2009). Across seven European countries, 25.13% of households included at least one grandparent in 2010 (29.70% in Eastern Europe, 5.46% in Western Europe); in contrast, skipped-generation families in this world region are rare (Masfety et al., 2019). In Asia, particularly China, an overwhelming number of grandchildren have been “left behind” in rural areas with their grandparents, while parents venture to cities in search of better employment opportunities (Ingersoll-Dayton et al., 2018; Knodel and Wassana, 2004; Snopkowski and Sear, 2015; Tang et al., 2018; Xu, 2019). In sub-Saharan Africa, there are also long traditions of grandparental care (Bentley and Mace, 2012; Bledsoe and Isiugo-Abanihe, 1989; Notermans, 2003), as well as a swell of grandparental involvement in the wake of HIV/AIDS epidemics and parental migration (Ice et al., 2012; Kasedde et al., 2014; Mazzucato et al., 2015). This literature, however, has not necessarily assessed the influence that grandparents have on the grandchildren. For example, some of the global literature on aging and health focuses on the positive or negative effects of care on the grandparents themselves (Courtin and Avendano, 2016; Hayslip et al., 2017; Kasedde et al., 2014; Komonpaisarn and Loichinger, 2019), while other literature has theorized grandparental investments in terms of lifetime reproductive fitness, to benefit elders (e.g., the “grandmother hypothesis”), explaining why post-reproductive women would promote the survival and nutrition of their grandchildren (Hawkes et al., 2017). Grandparents, thus, make major contributions to family and social life, although the care they provide varies in type, intensity, and whether they act as primary or secondary caregivers. Dunifon et al. (2014), for example, noted that “the population of children living with grandparents is very diverse in terms of family structure and economic wellbeing” (p.111), which carries different implications for the health of family members. Three-generation families in the U.S., “can look very different depending on the parent's marital or partnership status” (p.113), because children in two-parent, three-generation living arrangements are economically more advantaged than children living in single-mother, three-generation households, while children in grandparent custodial care are the most economically vulnerable. The comparison between three-generation and skipped-generation families, therefore, needs to account for socioeconomic contexts and rely on appropriate comparison groups. Studies drawing from national survey data have shown that teenagers living with single mothers and at least one grandparent, in multigenerational households in the U.S., have developmental outcomes that are “at least as good and often better than the outcomes of teenagers in married families” (Deleire and Kalil, 2002, p.393), perhaps because the grandparent(s) in Black, low-income families tend to be young and actively involved in caregiving. In the U.S., as well as in the U.K. and Australia, studies find that three-generation residence is more common for younger, less educated, and minority groups (Amorim et al., 2017; Pilkauskas and Martinson, 2014). The age of the child and the relationship status of the parent also affect this type of care (Kamo, 2000; Pilkauskas, 2012; Pilkauskas and Cross, 2018). In this systematic review, we sought to establish the global evidence base on the links between grandparents' involvement and grandchildren's health and development, in order to map what is known for skipped-generation and multigenerational families. We operationalized the parameters of grandparental involvement as follows: contact with grandchildren (co-residence, frequency of visits), caregiving behaviors (participation in child rearing), and support (such as investment of financial resources). This follows earlier literature seeking conceptual clarity with respect to the construct of grandfather involvement, which differentiated measures of “contact, participation in activities, and commitment” (Bates and Taylor, 2013). Two main research questions guided our review: how do grandparents influence child health and development, and what range of child outcomes is reported globally? We undertook a qualitative meta-synthesis of studies to date, rather than a meta-analysis of the strength of associations (effect sizes) between grandparental involvement and child health. Qualitative analyses are fitting where systematic reviews capture relatively new literature—such as research focusing on grandparent rather than parent caregivers—or capture a heterogeneous body of work, as they help generate questions focused on processes and theory development (Johnson and Hennessy, 2019). 2. Methods 2.1. Search strategies We followed “best practice recommendations” (Johnson and Hennessy, 2019) for research synthesis in the health sciences and methodological steps for systematic reviews, which synthesize the most important guidelines developed by AMSTAR 2 and PRISMA checklist items (Moher et al., 2009; Shea et al., 2017). We established a formal review protocol before starting our literature search. We built a research team with broad expertise in the social sciences, child health and development, and clinical medicine, to review the literature to date, identify research questions, and develop relevant search terms and inclusion/exclusion criteria. To define parameters for the review, we utilized elements captured by the PICOT and TOPICS + M acronyms (Johnson and Hennessy, 2019): we defined the period of time (T) for studies under review; specified that health and development outcomes (O) should pertain to grandchildren; selected a global coverage for populations (P) of interest; included intervention studies (I) that were relevant to parenting behaviors; noted appropriate comparison (C) groups for evaluating the roles and impacts of grandparents; decided not to limit the search by study design (S) or language of publication; and thought about a priori moderators (M) potentially influencing heterogeneity in the extant literature, such as age, income, race, place, crisis of separation, parent relationship status, and quality of relationships, in addition to types of co-residence. The review was registered in the PROSPERO database (CRD42019133894). We searched seven databases: MEDLINE, Embase, PsycINFO, Global Health, Anthropology Plus, Social Sciences Citation Index (Web of Science), and Sociological Abstracts (ProQuest). We piloted the search on March 16th, 2016, after which the research team worked to evaluate the scope and quality of the literature and to finalize the review protocol. A professional librarian ran the final systematic review search on May 12th, 2019, using keywords and controlled vocabulary for grandparents and children, and custodial or multigenerational care (Supplementary online Table 1). For example, the search captured 2 Region & country Study characteristics Child health and development outcome category 3 Study designb Sample size Sample characteristics AFRICA (n = 17) Cameroon MG CS 671 Ethiopia MG CS 700 The Gambia MG L Ghana MG CS All available records 2387 Kenya C CS 149 Kenya MG CS 509 Lesotho MG CS 10,641 Malawi MG CS 1635 Malawi MG CS 39 Adolescents in Bemenda, living in different household structures and of varying socioeconomic backgrounds Anthropometric survey and demographic data from children (≤16 yr) in the 1999 and 2003 censuses Records of residents of villages participating in medical surveys in the country's western region from 1950 to 1974 Adolescents (12–19 yr) participating in the 2004 Adolescent Survey Children (2–16 yr) with parent and grandparent caregivers Mothers, fathers and grandmothers participating in an intervention to promote optimal infant feeding practices through dialogue-based groups Children (6–16 yr) participating in the Lesotho Demographic and Health Survey Children from two villages located by Lake Malawi in the southern region of the country Clinical sample of children (4–6 yr) with obesity Nigeria MG CS 486 South Africa C CS 254 South Africa MG CS 3993 South Africa MG CS 512 South Africa MG CS 384 South Africa MG CS 120 Tanzania MG CS 3136 Uganda MG L 1940 ASIA (n = 68) Bangladesh MG CS Bangladesh MG E All available records 86 China C CS 6623 Parent and grandparent caregivers with children (3–24 mo) Children orphaned due to AIDS Infants with families participating in a biannual community survey (Africa Center Demographic Information System) Children (mean age 14 yr) from metropolitan Cape Town Physical health Socio-emotional and behavioral health Cognitive and educational development Rwenge, 2000 Sexual behavior Growth, mortality Gibson and Mace, 2005 Sear et al., 2002 Mortality Sexual behavior Tenkorang and Adjei, 2015 Oburu, 2005 Adjustment Mukuria et al., 2016 Infant feeding practices School enrollment Parker and Short, 2009 Mortality Sear, 2008 Infant feeding practices Infant feeding practices Kerr et al., 2007 Titiloye and Brieger, 2009 Sharer et al., 2015 Anxiety, depressive symptoms, post-traumatic stress symptoms, suicidal tendencies Cunningham et al., 2010 Birth weight Externalizing & internalizing behavior challenges, prosocial behavior, substance use Adjustment, prosocial behaviors Adolescents (8th & 9th graders) from metropolitan Cape Town Children (4th & 5th graders) from metropolitan Cape Town Profe and Wild, 2017 Levetan and Wild, 2016 Van Heerden and Wild, 2018 Externalizing & internalizing behavior challenges, prosocial behavior Children (< 5 yr) from 56 villages participating in the Whole Village Project Adolescent girls (16–19 yr) participating in the Rakai Community Cohort Study Height, weight Children born shortly before maternal death and children born immediately before the last child born before maternal death, from 1974 to 2005 Mothers with children (< 13 yr) from Matlab in various caregiving environments School-aged children, including left-behind children, in the Xiangxi region and Changsha city in Hunan province Mortality Lawson et al., 2017 Sexual behavior Pilgrim et al., 2014 Razzaque et al., 2014 Height, weight Educational attainment History of lifetime traumatic events, psychotic-like experience Perry, 2017 Sun et al., 2017 (continued on next page) Social Science & Medicine 239 (2019) 112476 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 Studies captured in the review, by geographic region, linking grandparent involvement to child outcomes (n = 206). Region & country Study characteristics Child health and development outcome category Sample size Sample characteristics China C CS 3510 Left behind adolescents (12–16 yr) from 16 rural high schools in Sichuan province, China China C CS 606 China C CC 886 China C CC 390 China MG L 1172 China MG L 317 China MG CS 15,054 Children (8–14 yr) from migrant worker families with kin and grandparent caregivers Children (1–3 yr) being fostered by grandparents, and age-matched children living with their parents, in Shanghai Pediatric clinic patients (1–14 yr), and matched controls, in rural Guigang Mother-infant dyads recruited from four public hospitals in Hong Kong Preschool children living with parents and grandparents in Guangzhou and Xi'an, China Children (2–13 yr) participating in the China Health and Nutrition Survey China MG CS 5142 China MG CS 2514 China MG CS 2117 Middle school students from Fuzhou City China MG CS 1410 Primary-school students in Changsha city BMI China MG CS 1383 Infant feeding practices China MG CS 1382 China China MG MG CS CS 1234 618 China MG CS 497 Caregivers of infants (18–30 mo) living in poverty in 351 villages in rural China Preschool children recruited from ten schools across six districts of Shanghai Preschool children in ten schools Children (6–16 yr) with chronic diseases from seven hospitals in Shanghai and Beijing Children (8–10 yr) from four primary schools in two southern cities in China China MG CS 428 China MG CS 422 China MG CS 287 China MG CS 227 China MG CS 99 Children (3–7 yr) participating in a national epidemiological survey Infants (12 mo) and their recruited from Peking University First Hospital Middle-school students, half of whom were left behind and cared for by grandparents Parents with children (3–6 yr) in various caregiving environments College students (mean age 19.79 yr) attending a university in Hong Kong Clinical sample of children with atopic dermatitis with parent, grandparent and other caregivers Stakeholders for children's health and weight status: parents, grandparents, teachers, nurses, and catering staff Physical health Socio-emotional and behavioral health Cognitive and educational development Bullying, depressive symptoms, panic & significant somatic symptoms, psychological distress, self-esteem Loneliness Tang et al., 2018 Psychological & behavioral development Wang et al., 2009 Jia and Tian, 2010 Risk of pesticide poisoning Infant feeding practices Li et al., 2011 Bai et al., 2016 Social competence Li and Liu, 2019 Eating behaviors physical activity, weight Obesity He et al., 2018 Zong et al., 2015 Occipital-frontal circumference, supine length for age, weight for age, weight for length Views of self Self-esteem Developmental delay Li et al., 2017a Association with peers, bonding Academic performance Liu, 2016 Yue et al., 2018 Aggressive behavior Jia et al., 2016 Aggressive behavior Behavioral challenges Jia et al., 2011 Zhang et al., 2013 BMI, eating behaviors, height, physical activity, weight Li et al., 2014 Life satisfaction, self-esteem Attachment, separation anxiety, temperament Breadth of friendships, depth of friendships, filial piety beliefs Traditional medicine utilization Obesity Xue-Yan et al., 2016 School engagement School adaptation Cognitive wellbeing Song et al., 2018 Sun and Jiang, 2017 Li et al., 2018 Hon et al., 2005 Li et al., 2015 (continued on next page) Social Science & Medicine 239 (2019) 112476 Study designb 4 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category Study designb Sample size Sample characteristics China MG CS 72 China MG CS 53 China China MG MG CS CS 26 23 China C& MG L China C& MG L All available records 146 Children (17.5 mo) cared for by grandmothers, while mothers were at work, in Beijing Parents and grandparents with dependent children (6–9 yr) in nuclear and three-generation households Children with parent and grandparent caregivers Parent and grandparent caregivers in three-generation households Children (2–12 yr) with families participating in the China Health and Nutrition Survey China C& MG C& MG C& MG C& MG C& MG C& MG C& MG C& MG CS 42,109 CS 9289 CS 4139 CS 2530 CS 2001 CS 1125 CS 946 CS 595 CS 480 CS 71 India C& MG C& MG MG CS 920 Students (mean age 13.97 yr) in rural junior and middle schools in Hunan Province Left-behind adolescents (12–16 yr) from Shandong province, whose fathers, or fathers and mothers, were rural-urban migrants Infants (25–30 mo) in urban areas, selected from a maternal-child care center birth cohort in Hefei Infants (mean age 17.6 mo), their mothers and their grandmothers Children (1–3 yr) living in peri-urban Punjab India MG CS 427 Adolescents in nuclear and three-generation households India MG CS 195 Lactating mothers enrolled in a community survey India MG CS 39 Indonesia MG DR 163 Rural families with children (3–24 mo), with and without grandmothers present Mother-infant dyads and grandmothers Japan MG L 43,046 Japan MG L 4281 China China 5 China China China China China China China Physical health Socio-emotional and behavioral health Freshman and sophomore college students (mean age 18.79 yr) attending a university in Shanghai College students (16–20 yr) attending a university in Shanghai during the 2014–2015 academic year Left-behind children in Hubei Province Children born in 2001 (followed from 2.5 to 13 yr) with families participating in the Longitudinal Survey of Newborns in the 21st Century Children (assessed at 3, 6 & 12 yr) enrolled in the Ibaraki Children's Cohort Xing et al., 2016 Socio-emotional development Eating behaviors, nutrition Nutritional status Eating behaviors, nutrition Overweight, underweight Goh Esther, 2013 Zhang et al., 2015 Jingxiong et al., 2007 de Brauw and Mu, 2011 Children (39–59 mo) in kindergartens in Beijing and Hangzhou Children (6–15 yr) living in Guangzhou or Shenzhen, two cities in Guangdong province New kindergarteners enrolled in school in Longhua New District of Shenzhen, China School children (mean age 6.3 yr) in Hong Kong Cognitive and educational development Intelligence, verbal ability, executive functioning, theory of mind Pedestrian injury Health-related quality of life Ma et al., 2010 Conduct behavior challenges Liu et al., 2018 Victimization and polyvictimization Perceived social support, selfesteem Self-esteem Chan et al., 2017 Eating behaviors Infant feeding practices Shi et al., 2017 Behavioral challenges, deliquent behaviors Depressive symptoms Zhao et al., 2016 Temperament Lu et al., 2012 Emotional and behavioral challenges, temperament Xing et al., 2016 Fan and Fang, 2010 Gurupdesh et al., 2013 Emotional intelligence BMI, eating behaviors Shi et al., 2017 Liu et al., 2019 Height for age, weight for age Infant feeding practices Infant feeding practices Infant feeding practices Overweight & obesity Guo, 2014 Renny and Jayasankara, 2016 Chaudhry et al., 2014 Sharma and Kanani, 2006 Susiloretni et al., 2015 Ikeda et al., 2017 Sata et al., 2015 (continued on next page) Social Science & Medicine 239 (2019) 112476 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category 6 Type of carea Study designb Sample size Sample characteristics Physical health Japan MG CS 53,575 MG CS 3832 Infant feeding practices Eating behaviors Kaneko et al., 2006 Japan Japan MG CS 3422 Infants enrolled in the First Longitudinal Survey of Babies in the 21st Century Study Young women (18–20 yr), their mothers an their paternal or maternal grandmother living in three-generation households Children (1st graders) and their caretaker living in Adachi City, Tokyo Morita et al., 2019 Japan MG CS 2381 BMI, eating behaviors, height, weight Mortality Japan MG CS 838 Japan MG DR 47,015 Nepal MG CS 1399 Nepal MG CS 211 Pakistan MG CS 99 South Korea C CS 680 South Korea MG CS 14,000 South Korea CS 414 Taiwan C& MG C CS 568 Taiwan MG CS 19,531 Taiwan MG CS 11,914 Taiwan MG CS 2510 Taiwan MG L 4971 Thailand C CS 320 Thailand MG CS 320 Thailand MG CC 232 Vietnam MG CS 1905 AUSTRALIA (n = 3) Australia C CS 20 Children in a population registry of mortality and household structure from 1671 to 1871 Clinical sample of children (mean age 12.4 yr) with H. pylori, with parent and grandparent caregivers Children with families participating in the Longitudinal Survey of Babies born in the 21st Century from 2001 to 2007 Infants and toddlers (16–24 mo) living with their grandparents Children born between July 1993 and June 1994 using data from the Nepal Demographic and Health Survey Family triads—mother, father and adolescent child (12–18 yr)—in three-generation households Young adolescents (11–15 yr) in four family structure groups, including grandparent families, in Seoul City and Kyunggi State School-aged children from a nationally representative sample, followed 2006–2012 Children in three different types of out-of-home care arrangements Dyads of adolescents (12–19 yr), infected with HIV via vertical transmission and their primary caregivers Children (7th graders) participating in the Taiwan Educational Panel Survey Children (7th graders) participating in the Taiwan Educational Panel Survey Children in early adolescence (7th–9th graders) Children (6–60 mo) treated at Siriraj Hospital and control cases matched for sex and age from the hospital's outpatient unit Children (12–16 yr) and adult caregivers of children ages 6–16 years Children (8–15 yr) with grandparent caregivers Cognitive and educational development Kobayashi et al., 2015 Jamison et al., 2002 Infectious disease transmission Accidental injury Morioka et al., 2018 Infant feeding practices Mortality Karmacharya et al., 2017 Li and Mora, 2016 Height Urita et al., 2013 Social skills Akhtar et al., 2017 Aggressive behavior Yun and Shin, 2013 Drinking behavior, smoking behavior Ko, 2019 School adjustment Adherence to medication Lee et al., 2017 Xu et al., 2017 Cognitive development Cognitive ability Life satisfaction, negative emotions, positive emotions Accidental poisoning Pong and Chen, 2010 Lin and Yi, 2018 Educational attainment Intellectual development Non-verbal intelligence Nutritional status Chen, 2016 Chiang and Park, 2015 Nanthamongkolchai et al., 2011 Nanthamongkolchai et al., 2009 Sutchritpongsa, 2016 Externalizing & internalizing behavior challenges Weiss et al., 2014 Emotional wellbeing, experience of grandparent care, self-concept Downie et al., 2010 (continued on next page) Social Science & Medicine 239 (2019) 112476 Children (surveyed annually beginning in 7th or 9th grade) participating in the Taiwan Youth Project Children (6–12 yr) with parent and grandparent caregivers Children (6–12 yr) living in Phrae Province Socio-emotional and behavioral health Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category 7 Type of carea Study designb Sample size Sample characteristics Physical health Australia MG CS 114 Eating behaviors Rhodes et al., 2016 Australia MG CS 11 Members of 27 three-generation families of Anglo-, Chinese- and Italian-Australian descent Grandparents of young children (1–5 yr) Eating behaviors Rogers et al., 2019 EUROPE (n = 32) Bulgaria MG CS 62 Denmark MG CS 2383 England England MG MG CS CS 100 30 England MG CS 9 England MG CS 9 Finland MG CS 71,551 Germany MG L Greece MG CS All available records 729 Hungary Ireland MG MG CS L 509 11,314 Ireland MG CS 10,748 The Netherlands MG CS 1540 Spain C CS 70 Spain MG CS 32,234 Spain MG CS 32,234 Sweden MG CS Sweden MG CS All available records 11,768 Sweden MG CS 7719 Sweden MG CS 39 Adolescents (mean age 14.2 yr) with parent and grandparent caregivers Children who participated in the Danish Longitudinal Survey of Youth Parent and grandparent caregivers Grandmothers participating in a focus group; mothergrandmother pairs in an intervention Professionals working with young mothers participating in a study about infant feeding practices among their clients Families with young children (3–6 yr) of Bangladeshi or mixed-ethnicity speaking Sylheti and Bengali, or English, respectively Children with families participating in the Finnish Census Panel Genealogies from German villages from 1412 to 1912 Children (9–13 yr) in a representative sample Children (10–14 yr) and their parents and grandparents Children (assessed at 9 mo and at 3 yr) with families participating in the infant cohort of the Growing Up in Ireland study Singleton babies (9 mo) and primary caregivers enrolled in the Growing Up in Ireland Study Cognitive and educational development Depressive symptoms Choice of secondary education Eating behaviors Infant feeding practices Infant feeding practices Botcheva and Feldman, 2004 Møllegaard and Jæger, 2015 Farrow, 2014 Ingram et al., 2003 Bernie, 2013 Language development, literacy Academic performance Jessel et al., 2011 Lehti et al., 2019 Mortality Kemkes-Grottenthaler, 2005 Overweight & obesity Physical activity Moschonis et al., 2010 Lakó, 2014 McDonnell, 2016 Emotional & behavioral challenges Motor development Children participating in the Longitudinal Internet Studies for Social Sciences survey Children being fostered by grandparents in Málaga, Spain Social development Language development, problem-solving skills Educational attainment Externalizing & internalizing behavior challenges Smoking behavior BMI, height, weight Bol and Kalmijn, 2016 Fuentes and Bernedo, 2009 Escario and Wilkinson, 2015 Duarte et al., 2016 Cigarette consumption, smoking behavior BMI Cruise and O'Reilly, 2014 Academic performance Hallsten and Pfeffer, 2017 Academic performance Cognitive ability Modin et al., 2012 Modin and Fritzell, 2009 Lindberg et al., 2016 (continued on next page) Social Science & Medicine 239 (2019) 112476 Students (14–18 yr) participating in the 2010 State Survey on Drug Use in Secondary School Students Students (14–18 yr) participating in the 2010 State Survey on Drug Use in Secondary School Students Children (9th graders) born from 1980-1996 with traceable grandparents and whose information was included in national registries Grandchildren (9th graders) of individuals born in Uppsala from 1915 to 1929 Young men (mean age 18.2 yr) conscripted into the Swedish military from 1980 to 2002 Clinical sample of children (4–6 yr) Socio-emotional and behavioral health Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category 8 Study designb Sample size Sample characteristics United Kingdom MG L 13,744 United Kingdom MG L 12,319 United Kingdom MG L 9000 United Kingdom MG L 8752 United Kingdom MG L 4800 United Kingdom MG CS 18,827 United Kingdom MG CS 15,109 United Kingdom MG CS 1596 United Kingdom MG CS 1488 United Kingdom MG CS 801 United Kingdom MG CS 47 United Kingdom MG CC 200 Children (assessed at 9 mo & 5 yr) enrolled in the British Millennium Cohort Study Children (assessed at 9 mo & 3 yr) enrolled in the Millennium Cohort Study Children (assessed at 3, 5 & 7 yr) participating in the British Millennium Cohort Study Families with children (4 yr) enrolled in the Avon Longitudinal Study of Parents and Children Children (3 yr) enrolled in the Millennium Cohort Study whose mothers were working when the children were 9 mo Infants (9 mo) enrolled in the Millennium Cohort Study (data from the first wave) Families with children (3 yr) enrolled in the Millennium Cohort Study Children (11–16 yr) attending one of 70 secondary schools across the UK Children (11–16 yr) who participated in the Involved Grandparent and Child Well-Being survey in 2007 in England and Wales Children (11–16 yr) attending one of 68 secondary schools across England and Wales Families residing in Cambridgeshire, England from 1770 to 1861 Children (2–5 yr) fed by their mothers or grandmothers LATIN AMERICA (n = 12) Brazil MG L 601 Brazil MG L 323 Brazil MG L 323 Brazil MG L 220 Brazil MG CS 91 Brazil MG RCT 492 Brazil MG RCT 323 Brazil C& MG MG CS 211 L 15,175 Chile Mothers who gave birth to healthy babies at a university general hospital in Porto Alegre Adolescent mothers who gave birth at the rooming-in facility of Hospital de Clínicas de Porto Alegre, their mothers, and their infants Adolescent mothers who gave birth at the rooming-in facility of Hospital de Clínicas de Porto Alegre, their mothers, and their infants Dyads—mothers and infants (assessed at 7 days, and 1, 2, 4 & 6 mo)—recruited in a hospital maternity ward in Porto Alegre Grandparents of children (6 mo–2 yr) in São Paolo who provide support at least weekly Parent and grandparent caregivers Adolescent mothers (< 20 yr) living with their infants, half of whom were also living with their own mothers and half of whom were not Mothers and their infants (1 mo) living in Porto Alegre, Brazil Children (0–5 yr) participating in a nationally representative survey Physical health Socio-emotional and behavioral health Cognitive and educational development Emotional & behavioral challenges Cognitive development Accidental injury Cognitive development Emotional & behavioral challenges, prosocial behaviors Emotional & behavioral challenges Tanskanen and Danielsbacka, 2018 Tanskanen and Danielsbacka, 2017 Del Boca et al., 2018 Fergusson et al., 2008 Language development, school readiness Infant feeding practices Overweight & obesity Hansen and Hawkes, 2009 Emmott and Mace, 2015 Tanskanen, 2013 Adjustment Griggs et al., 2010 Emotional & behavioral challenges Tanskanen and Danielsbacka, 2012 Prosocial behavior, psychopathology Flouri et al., 2010 Inter-birth interval, mortality BMI, height, weight Ragsdale, 2004 Metbulut et al., 2018 Infant feeding practices BMI-for-age zscore, height for age, infant feeding practices Eating behaviors, infant feeding practices Infant feeding practices Susin et al., 2005 Infant feeding practices Infant feeding practices Infant feeding practices Ferreira et al., 2018 Schwartz et al., 2015 Soldateli et al., 2016 Giugliani et al., 2008 Nunes et al., 2011 Giugliani et al., 2019 Infant feeding practices Franca et al., 2008 Language development Reynolds et al., 2018 (continued on next page) Social Science & Medicine 239 (2019) 112476 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category 9 Study designb Sample size Sample characteristics Physical health Guatemala MG CS 6262 Height for age Mexico MG L 4649 Mexico MG L 2442 Children and their mothers participating in the Guatemalan Survey of Family Health Children (3–12 yr) participating in the Mexico Family Life Survey Children (born 1991–1993) participating in the Mexico Family Life Survey MIDDLE EAST (n = 2) Israel MG CS 2751 Israel CS 1236 UNITED STATES (n = 60) United States C L United States C L All available records 10,083 United States C L 7844 United States C L 2402 United States C L 301 United States C CS 79,960 United States C CS 67,588 United States C CS 10,611 United States C CS 733 United States C CS 733 United States C CS 343 United States C CS 230 United States C CS 144 United States C CS 84 MG Socio-emotional and behavioral health Cognitive and educational development Sheppard and Sear, 2016 Schmeer, 2013 Iron-deficiency anemia Educational attainment, school transition Arab and Jewish middle- and high-school students (12–18 yr, 7th–11th graders) Children (grades 8–12) residing in a residential care setting Attar-Schwartz and Khoury-Kassabri, 2016 Attar-Schwartz and Huri, 2019 Adjustment, prosocial behavior Life satisfaction Children (6–17 yr) with families participating in the National Survey of America's Families in the 1999 and 2002 waves Adolescents (7th–12th graders) enrolled in the National Longitudinal Study of Adolescent Health, living in one of eight caregiving environments Adolescents (12–18 yr) participating in the National Longitudinal Study of Adolescent Health Children (10–14 yr) from low-income, urban neighborhoods participating in a household-based, stratified random sample survey Children (mean age 9.44 yr) with parent and grandparent caregivers Children (mean age 14 yr) involved with the Department of Juvenile Justice Children (0–17 yr) participating in the National Survey of America's Families Caregivers and their children (4–17 yr) enrolled in an NIMH study or participating in the 2001 National Health Interview Survey Grandparent caregivers of children (3 mo–16 yr) with parents absent for 3+ months Grandparent caregivers of children (3 mo–6 yr) with parents absent for 3+ months Children (4–12 yr) with grandparent caregivers enrolled in an RCT for improving family well-being Children (2–16 yr) born to adolescent parents with grandparent primary caregivers Current or former grandparent caregivers Health status, physical health condition Children (9 yr) enrolled in the Fragile Families Child Wellbeing Study birth cohort with grandparent caregivers BMI, physical health condition Physical health condition Behavioral challenges, mental health condition Arenas, 2017 Learning disability, school engagement Educational attainment Conway and Li, 2012 Monserud and Elder, 2011 Behavioral challenges Goulette et al., 2016 Externalizing & internalizing behavior challenges Pittman, 2007 Psychological wellbeing Hayslip et al., 2014 Substance use Shaw et al., 2019 Mental health condition Ziol-Guest and Dunifon, 2014 Smith et al., 2018 Externalizing & internalizing behavior challenges Adjustment Smith et al., 2008 Externalizing & internalizing behavior challenges Emotional & behavioral challenges Externalizing & internalizing behavior challenges Smith et al., 2015 Smith and Palmieri, 2007 Kelley et al., 2011 Accidental injury Externalizing & internalizing behavior challenges, mental health condition, social skills Academic performance, cognitive ability, developmental delay Nadorff and Patrick, 2018 Pilkauskas and Dunifon, 2016 (continued on next page) Social Science & Medicine 239 (2019) 112476 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category 10 Study designb Sample size Sample characteristics United States C CS 79 United States United States C C CS CC 14 108 United States C CC 108 United States C RCT 343 United States C RCT 14 United States C DR 66 Families with children (3–7.5 yr) and grandparent caregivers involved with a state prison system or local community agency Children (10–16 yr) with grandparent caregivers African American children (mean age 9.6 yr) with parent and grandparent caregivers African American children (mean age 9.6 yr) with parent and grandparent caregivers Grandmothers and their grandchildren (4–12 yr) from 4 different states Dyads—grandmothers or great-grandmother and children (2–7 yr)—enrolled in an RCT to evaluate a Child Directed Interaction Training program for children in kinship care Adolescents (12–18 yr) involved with the juvenile justice system, with parent and grandparent caregivers United States MG L 12,303 United States MG L 8650 United States MG L 8600 United States MG L 6550 United States MG L 6501 United States United States MG MG L L 3449 2846 United States MG L 2041 United States MG L 398 United States MG L 397 United States MG L 324 United States MG L 316 United States MG L 217 United States MG L 204 United States MG L 170 Mothers enrolled in the Early Childhood Longitudinal Study (8250) or in the Fragile Families and Child Wellbeing Study (4053) Infants (assessed at 9 & 24 mo) enrolled in the Early Childhood Longitudinal Study Children (assessed at 9 & 24 mo) enrolled in the Early Childhood Longitudinal Study Children from three-generation households with families enrolled in the Early Childhood Longitudinal Study Children from single-mother and three-generation households participating in the National Longitudinal Survey of Youth Children (0–33 mo) enrolled in the Healthy Steps Study Adolescents (14–17 yr) participating in the National Longitudinal Survey of Youth Mothers (assessed 3rd trimester–1 yr postpartum) enrolled in the Infant Feeding Practices Study II American Indian adolescents (7th–11th graders) participating in the National Longitudinal Study of Adolescent Health Mothers enrolled in the Michigan Bone Health and Metabolism Study Mother-infant dyads from a hospital sample recruited shortly after birth Grandparent-children (12th graders) dyads enrolled in the Iowa Youth and Families Study Families with children (0–12 yr) enrolled in the Panel Study of Income Dynamics Adolescent mothers of Mexican-origin and their young children Latina mothers and toddlers (12–24 mo) Physical health Socio-emotional and behavioral health Cognitive and educational development Poehlmann et al., 2008 Attachment, externalizing & internalizing behavior challenges Experience of grandparent care Externalizing & internalizing behavior challenges Emotional and behavioral challenges Externalizing & internalizing behavior challenges Externalizing & internalizing behavior challenges Physical health condition Mental health condition, peer relationships, risk of juvenile reoffending, substance use Kelch-Oliver, 2011 Edwards, 2006 Edwards, 2009 Smith et al., 2018b N’Zi et al., 2016 School challenges Infant feeding practices Campbell et al., 2006 Pilkauskas, 2014a Behavioral development Externalizing & internalizing behavior challenges, prosocial behaviors Cognitive development Cognitive development Academic performance Cognitive development, stimulation Accidental injury Sun and Li, 2013 Mollborn et al., 2011 Pilkauskas, 2014b Dunifon and Kowaleski-Jones, 2007 Bishai et al., 2008 Harper and McLanahan, 2004 Odom et al., 2014 Incarceration Infant feeding practices Eitle et al., 2014 Delinquent behaviors Birth weight Rillamas-Sun et al., 2014 Brenner et al., 2001 Immunization status Grandparent-grandchild relationship quality School readiness Compliance, externalizing behavior challenges Cognitive & language development Crosnoe and Elder, 2002 Augustine and Raley, 2013 Seay et al., 2016 Grau et al., 2015 (continued on next page) Social Science & Medicine 239 (2019) 112476 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category 11 Study designb Sample size Sample characteristics United States MG L 133 United States MG CS 198,844 Adolescent mothers of Mexican-origin and their young children Children (0–17 yr) participating in the National Health Interview Survey United States MG CS 29,345 United States MG CS 17,101 United States MG CS 12,029 United States MG CS 6300 United States MG CS 792 United States MG CS 700 United States MG CS 594 United States MG CS 551 United States MG CS 310 Parents attending WIC clinic in Washington, D.C. United States MG CS 283 United States MG CS 254 United States MG CS 200 Genealogical data and vital statistics records of children in Utah frontier families who founded a commune in 1874 Toddlers with parent and grandparent caregivers from a randomly selected representative community sample from upstate NY Children (10–69 mo) from families who called regional poison control centers United States MG CS 199 Children (5–12 yr) from families of Hispanic descent United States MG CS 194 United States MG CS 187 United States MG CS 126 United States MG CS 53 United States MG RCT 248 Mothers and children (4–5 yr) enrolled in the LONGSCAN Study Children (10–14 yr) treated in pediatric emergency departments at a hospital in Philadelphia or a hospital in Baltimore Parent- infant (0–6 mo) dyads presenting for well-child visits at Children's National Medical Center, Washington, D.C. Latinx grandparents caring for or providing support to their grandchildren (2–12 yr) Women and infants enrolled in an RCT of an intervention to promote safe sleep in infants Children (5th, 8th, 9th & 11th graders) from White, Somali, Hmong and Latinx families, living in one of six caregiving environments Adolescents from three-generation households enrolled in the National Longitudinal Study of Adolescent Health Chinese-American children (2–19 yr) being cared for by at least one grandparent Children with families participating in the Early Childhood Longitudinal Study, 2001 Birth Cohort Children enrolled in the Rochester Intergenerational Study Children entering kindergarten (in Fall 2006 or Fall 2007) who are members of the Early Childhood Longitudinal Study 2001 birth cohort Mothers who had breastfed, or were currently breastfeeding, a child when surveyed Children (14–19 yr) participating in the National Survey of Families and Households Physical health Physical health condition Physical health condition Socio-emotional and behavioral health Attention Deficit Disorder, Attention Deficit/Hyperactivity Disorder Mental health condition, substance use & exposure Cognitive and educational development Effortful control, school readiness Learning disability, school days missed Jahromi et al., 2018 Krueger et al., 2015 Areba et al., 2018 Challenging behaviors, depressive symptoms Hamilton, 2005 Weight Lau et al., 2019 Emotion regulation Noah, 2018 Substance use Thornberry et al., 2006 Cognitive development Infant feeding practices Fomby et al., 2015 Cisco, 2017 Grandparent-grandchild relationship quality, risky behaviors, sexual behavior Academic performance Infant sleeping position Mortality Dunifon and Bajracharya, 2012 Moon et al., 2004 Heath, 2003 Brook et al., 2002 Challenging behaviors Potential access to grandparents' medication BMI, eating behaviors, physical activity McFee and Caraccio, 2006 Pulgaron et al., 2016 Externalizing & internalizing behavior challenges Cognitive development Black et al., 2002 Violence-related injury Ryan et al., 2019 Infant sleeping position Moon and Omron, 2002 Physical activity Xie et al., 2018 Infant sleeping position Flick et al., 2001 (continued on next page) Social Science & Medicine 239 (2019) 112476 Type of carea Reference A.F.A. Sadruddin, et al. Table 1 (continued) Region & country Study characteristics Child health and development outcome category Type of carea Study designb Sample size Sample characteristics Physical health United States MG RCT 181 Infant feeding practices United States C& MG L 2402 United States C& MG C& MG CS 5962 CS 157 Mothers and grandmothers enrolled in an RCT of an intervention to delay the introduction of complementary foods in infancy Children (2–4 yr) from low-income, urban neighborhoods participating in a household-based, stratified random sample survey Children living with at least one foreign-born parent or grandparent in the household Grandchildren (8–9 yr) of Ohio grandmothers participating in a National Institutes of Health study of grandparents as caregivers Height United States MULTIPLE COUNTRIES (n = 12) Brazil & Peru MG CS 38,327 Academic performance Pittman and Boswell, 2007 Minority language maintenance Ishizawa, 2004 Depressive symptoms, resourcefulness CS 69,125 China & United States C CS 344 Europe & Israel MG CS 24,447 Europe MG CS 4582 South Africa & United States C CS 952 Sub-Saharan Africa Sub-Saharan Africa 8 European Countries MG CS 917,788 MG CS 344,748 Children (6–60 mo) in three-generation households Stunting MG L & CS 12,350 Children (mean age 7.8 yr in cross-sectional survey and mean age 5.9 yr in longitudinal analysis) in extended household structures BMI, sum of tricipital & subscapular skinfold thickness, waist-toheight ratio 33 OECD Countries 41 OECD Countries 41 OECD Countries MG CS 259,625 MG CS 193,841 MG CS 107,834 Children (15–16 yr) participating the in the 2012 Program for International Student Assessment Students (15 yr) participating in the 2000 Program for International Students Assessment Students (15 yr) participating in the 2000 Program for International Students Assessment Musil et al., 2018 Academic performance Mortality Pérez et al., 2007 Dong et al., 2017 Adjustment, interpersonal dynamics Wang et al., 2019 Educational attainment Externalizing & internalizing behavior challenges Bruederle et al., 2019 Sexual behavior Locus of control Deindl and Tieben, 2017 Masfety et al., 2019 School enrollment Schrijner and Smits, 2018a Schrijner and Smits, 2018a Formisano et al., 2014 Academic performance Academic performance Academic performance Radl et al., 2017 C, custodial care (grandparents were primary caregivers of children, e.g., in skipped-generation families); MG, multigenerational care (both grandparents and parents were caregivers). L, longitudinal; CS, cross-sectional; CC, case-control; RCT, randomized controlled trial; DR, descriptive retrospective; E, ethnographic. Chiu and McBrideChang, 2010 Chiu, 2007 Social Science & Medicine 239 (2019) 112476 MG 12 China, Japan & Taiwan b Cognitive and educational development Black et al., 2001 Externalizing & internalizing behavior challenges, selfregulation Children in Brazil and Peru living with or without grandparents in the household Boys (1–9 yr) living in various household arrangements in historical populations from northeast China (1789–1909), northeast Japan (1716–1870), and north Taiwan (1906–1945) Grandparents and children in skipped generation households (China and US) or other arrangements in which grandparents raised grandchildren without assistance of parents (US only) Adults (50+ yr) participating in the Survey of Health, Aging and Retirement in Europe Children from six western and eastern European countries participating in the School Children Mental Health in Europe survey Adolescents (15–19 yr) participating in the global Wellbeing of Adolescents in Vulnerable Environments study in Baltimore, Maryland, USA and Johannesburg, South Africa Children (7–15 yr) in three-generation households a Socio-emotional and behavioral health Reference A.F.A. Sadruddin, et al. Table 1 (continued) Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. Fig. 1. Literature search PRISMA flow diagram. identifying a total of 1,041 publications for full-text assessment. We then applied three inclusion criteria to full-text screening: studies that (a) reported outcomes for grandchildren from 0 to 21 years of age, (b) included indicators of grandparental contact, caregiving behaviors, and/or support, and (c) were peer-reviewed publications. We excluded (a) 543 studies that reported outcomes specific to grandparents, rather than grandchildren, and 51 studies that focused on outcomes of grandchildren over 21 years old; (b) 179 studies that did not link indicators of grandparent involvement to child outcomes, beyond mentioning the influence of grandparents in framing abstracts; and (c) 62 studies that were non peer-reviewed, such as dissertations and book chapters. We used a standardized inventory to code the descriptive and methodological features of studies for purposes of descriptive categorization and assessment of methodological quality. Using the GRADE approach as a guide, we systematically reviewed the quality of evidence for potential areas of bias among individual studies. We focused on how vocabulary terms such as grandmother, grandfather, elder; infant, teen; adoption, caring, raising, foster care; blended family, extended family, co-residence, skipped-generation, three-generation, multigenerational. Specific search parameters were adjusted, where necessary, for database-appropriate syntax. Citations from all databases were imported into an EndNote X9 library. The duplicate citations were removed within Endnote. The database of citations with abstracts was then uploaded into Covidence, a screening and data extraction tool. 2.2. Screening eligibility and inclusion criteria The search yielded 23,567 records. We excluded duplicates (n = 7,148), and records published prior to 2000 (n = 3,720) because this literature rarely focused on the grandchildren's outcomes. Two authors independently examined 12,699 abstracts, in order to exclude those that did not explicitly report child outcomes (n = 11,658), thus 13 Country Study details Australia 20 children, 8–15 yr, living full Children from white Caucasian and time in the care of grandparents aboriginal families cared for full time by grandparents for periods of 1–12 years, with family histories of parental substance use, mental health and complications, and maternal death Cross-sectional analysis of 606 children, 8–14 yr (324 from Relationships, frequency of loneliness in children left with migrant worker families), raised communication with parents, and grandparents by kin or grandparents economic status vary among children left behind to be raised by grandparents Case control study of risk factors 390 pediatric clinic patients A range of risk and protective for pesticide poisoning (1–14 yr), and matched factors may contribute to poisoning controls, in rural Guigang risk, including safety of the home environment and how children are educated and looked after with regard to avoiding dangerous substances in the home Left-behind children can Cross-sectional analysis of risk 6623 school-aged children, including left-behind children, factors for psychotic-like experience a range of adverse in the Xiangxi region and experiences of left-behind effects due to separation from their Changsha city in Hunan children parents province China China China Sample characteristics Context of care Cross-sectional analysis of resilience and wellbeing of children living with grandparents 14 Outcome categorya Outcome(s) Key findings SB Emotional wellbeing, experience of grandparent care, self-concept Children raised by grandparents have Downie et al., 2010 appropriate levels of self-worth and emotional health; themes used to describe caregiving setting include protective factors, risk factors and coping strategies SB Loneliness Left-behind children are 2.5 times more likely to experience loneliness and 6.4 times more likely to report feeling very lonely Jia and Tian, 2010 PH Risk of pesticide poisoning Risk of pesticide poisoning is higher in families where grandparents are the custodial caregivers Li et al., 2011 SB History of lifetime traumatic events, psychotic-like experience Sun et al., 2017 SB Bullying, depressive symptoms, panic & significant somatic symptoms, psychological distress, self-esteem Left-behind children are more likely to experience psychotic-like events than their peers, but having a grandparent as a primary caregiver is associated with less severe suffering from these events Rates of low self-esteem and depression increase with length of separation from parents, but this risk is reduced when grandparents are primary caregivers Chinese grandparents report greater emotional and behavioral problems of grandchildren than American grandparents, which may be related to cultural differences in stigma attached to being raised by a grandparent Among the Chinese sample, attachment is negatively associated children's interpersonal difficulties Cross-sectional analysis of mental health and psychosocial problems in left-behind children 1663 left-behind children and 1683 controls, 12–16 yr Many left-behind children experience mental health and psychosocial problems China Case control study of crosscultural differences in custodial grandparent care and grandchildren's wellbeing 886 children, 1–3 yr, being fostered by grandparents, and age-matched children living with their parents, in Shanghai Cultural factors can account for SB differences in a variety of aspects of grandparental care and associated outcomes among grandchildren China & United States Cross-sectional analysis comparing Chinese and American caregiving grandparents' challenges and resources 238 grandparent caregivers from the US and 106 from China, and their grandchildren Kenya Cross-sectional analysis of custodial grandparent care and orphans' adjustment South Africa Cross-sectional analysis of caregivers and orphans to identify protective factors for good mental health Psychological & behavioral development In the American sample, SB grandparents raised grandchildren without the assistance of family in the parent generation, while in the Chinese sample, all participants were skipped-generation families HIV/AIDS contributed to the SB 34 orphans, 2–16 yr, with custodial grandmothers and 115 prevalence of skipped-generation children with biological parents families consisting of grandparents and orphaned children Adjustment, interpersonal dynamics 254 children orphaned due to AIDS Anxiety, depressive symptoms, post-traumatic stress symptoms, suicidal tendencies Orphans were cared for by parents, SB grandparents, or other kin, in densely populated neighborhoods formerly designated for Black Africans under apartheid, with high Adjustment Tang et al., 2018 Wang et al., 2009 Wang et al., 2019 Children raised by grandmothers are Oburu, 2005 similar to those raised by biological mothers with regard to adjustment, although grandmother caregivers report greater stress Children living with a grandparent or Sharer et al., 2015 other non-parent caregiver (kin/nonkin) have lower anxiety symptoms than those children living with a biological parent (continued on next page) Social Science & Medicine 239 (2019) 112476 China Reference A.F.A. Sadruddin, et al. Table 2 Studies of grandparent custodial care in skipped-generation families (n = 35, total sample). Country Study details South Africa & Cross-sectional analysis of the United effects of social support and States living arrangements on adolescents' risk-taking behaviors Sample characteristics Context of care rates of violence, unemployment, HIV prevalence (23–30%), and poverty Social support given by primary 456 adolescents living in caregivers may differ across Baltimore and 496 living in cultures even in contexts with Johannesburg, 15–19 yr, participating in the Wellbeing of similar rates of poverty, violence, and unstable living environments Adolescents in Vulnerable Environments Study Outcome categorya Outcome(s) Key findings Reference SB Sexual behavior Living with a grandparent may reduce teens' sexual risk-taking behaviors; specifically, Baltimore girls raised by a grandmother are less likely to be sexually active and Johannesburg boys raised by a grandfather are more likely to use condoms Factors that predict adolescent aggression vary by family structure; aggression in grandparent families is predicted by depression, anxiety, and level of family support Most children fall within the normal range of behavioral challenges; adolescents report fewer behavioral problems when grandparents use an inductive parenting style Adolescents with grandparents as the primary caregiver are less likely to adhere to optimal treatment standards Bruederle et al., 2019 Children raised by grandparents are more likely to have below-normal development and to experience inappropriate child rearing; income level affects child development for those raised by grandparents Youth in grandparent-headed homes are more likely to have: prior serious adjudication, a first delinquent offense by age 12, a history of associating with delinquent peers, prior assaults, undisciplined/ delinquent referrals, school problems, and parents/guardians unable or unwilling to provide supervision Children in grandparent families have the worst reported health statuses; grandparents provide greater access to health insurance and care but fewer extra-curricular activities; few significant differences are found for child outcomes between grandparent families and other nonparent family structures Children raised by grandparents show greater levels of externalizing and internalizing behavior Nanthamongkolchai et al., 2011 680 young adolescents (11–15 yr) in four family structure groups, including grandparent families, in Seoul City and Kyunggi State 70 children being fostered by grandparents in Málaga, Spain SB Adolescent aggression is of concern, given the increasing crime rates reported in Korea among early adolescents 568 dyads of adolescents, 12–19 yr, infected with HIV via vertical transmission, and their primary caregivers Children and adolescents PH transitioning into adulthood may face challenges assuming responsibility over their treatment adherence Most grandparents raising children PH; CE were not educated, had insufficient income, and experienced one or more family crises Spain Cross-sectional analysis of behavioral challenges of adolescents being cared for by grandparents Taiwan Cross-sectional analysis of risk factors for anti-retroviral treatment adherence of HIV positive youth Thailand Cross-sectional analysis of health 320 children, 6–12 yr, living status of children raised by with parents or grandparents parents vs. grandparents United States Descriptive retrospective study of reoffending risk for youth in grandparent- vs. parent-headed homes 29 children, 12–18 yr, in grandparent-headed homes and 37 children, of the same age range, in parent-headed homes, with juvenile justice system involvement Participants resided in an urban community in North Carolina and more than 80% were from racial minority backgrounds (primarily African American) PH; SB; CE Physical health condition; mental health condition, peer relationships, risk of juvenile reoffending, substance use; school challenges United States Longitudinal study of associations between family structure and child physical, behavioral, and educational outcomes All records available for children, 6–17 yr, with families participating in the 1999 and 2002 waves of the National Survey of America's Families Family structure and gender of the primary caregiver may be related to availability of resources and decision-making processes in the household PH; SB; CE Health status, physical health condition; behavioral challenges, mental health condition; learning disability, school engagement United States Case-control study of socialemotional behavior of children 54 African-American children raised by grandparents and 54 Participants selected from schools in a large, urban school district in the southeastern U.S., in SB Externalizing & internalizing behavior challenges Parenting styles among grandparents may influence adolescents' perceptions of their behavioral challenges SB Aggressive behavior Externalizing & internalizing behavior challenges Adherence to medication Nutritional status; intellectual development Yun and Shin, 2013 Fuentes and Bernedo, 2009 Xu et al., 2017 Campbell et al., 2006 Conway and Li, 2012 Edwards, 2006 (continued on next page) Social Science & Medicine 239 (2019) 112476 Cross-sectional analysis factors contributing to adolescent aggression among youth in different family structures 15 South Korea A.F.A. Sadruddin, et al. Table 2 (continued) Country Study details Sample characteristics Context of care raised by grandparents vs. biological parents matched children raised by parents, mean age 9.6 yr United States Case-control study of teachers' perceptions of serious emotional and behavioral problems in children in different family structures 54 children in grandparentheaded homes and 54 matched children living with their biological parents, mean age 9.6 yr communities where many residents had low socio-economic status (e.g., qualifying for free/reduced school lunch) Participants selected from schools SB in a large, urban school district in the southeastern U.S., in primarily African American communities where many residents had low socio-economic status United States Longitudinal study of adolescent 7844 children, 12–18 yr, antisocial and problem behavior participating in the National in custodial grandparent families Longitudinal Study of Adolescent Health 79 grandparents and 222 Longitudinal study of biological parents of children, grandparent-grandchild mean age 9.44 yr wellbeing United States United States 16 United States Custodial grandmothers may be able to provide more supervision and support than some biological or foster parents Caregivers had a range of backgrounds in terms of racial/ ethnic identity, marital status, employment status and annual income Outcome categorya SB SB Cross-sectional analysis of 14 children, 10–16 yr, and their Participants were African SB experiences of African-American grandparent caregivers American, and grandchildren had children raised by grandparents lived in grandparents' homes for at least 2 months for reasons including parents' medical illness, death, or mental or psychiatric illness; teen pregnancy; eviction or housing problems; neglect; or substance use Participants resided in one of two SB Cross-sectional analysis of 230 children, 2–16 yr, from baseline data of intervention to households with absent parents, major metropolitan areas in the southeastern U.S., with some where grandmothers are the improve grandparental grandchildren raised by primary caregivers wellbeing grandparents through formal custody agreements, and others not Longitudinal study of relationships between household structure & educational attainment of children 10,083 children, in grades 7 to 12, living in one of eight different caregiving environments, enrolled in National Longitudinal Study Adolescent Health United States Randomized controlled trial to improve outcomes for children living with grandmothers or great-grandmothers United States Cross-sectional analysis of grandchild wellbeing, with 14 dyads consisting of a Children predominantly Caucasian, SB grandmother or greatalthough African, American, grandmother and a child, 2–7 yr Hispanic and biracial children also included; caregivers came from an array of socioeconomic and educational backgrounds Caregivers were predominantly PH married, employed, and Caucasian CE Key findings Reference challenges in comparison with children living with biological parents Edwards, 2009 Emotional & behavioral challenges Teachers rate children raised by grandparents as exhibiting higher levels of a specific pattern of behavioral challenges, including somatic complaints, anxiety, depression, social challenges, and thought and attention difficulties Behavioral challenges Caregiver type does not significantly Goulette et al., 2016 predict adolescent problem behaviors Psychological wellbeing Experience of grandparent care Externalizing & internalizing behavior challenges Educational attainment Externalizing & internalizing behavior challenges Accidental injury Grandchild and grandparent Hayslip et al., 2014 wellbeing are related in a bidirectional manner, but grandchild difficulties have a greater impact on grandparent distress than do grandparent difficulties on grandchild distress Kelch-Oliver, 2011 Children experience adjustment issues as a result of family disruption and parental loss, but generally feel content living with their grandparents Children of grandmother caregivers who have increased psychological distress, fewer family resources, less social support, and less supportive home environments have increased externalizing and internalizing behavioral challenges Children in skipped-generation households have persistent disadvantages in terms of educational attainment: they are less likely to complete high school and enroll in or complete college (this is partially explained by socioeconomic resources, but not parenting quality) The CDIT intervention protocol results in significantly decreased internalizing behaviors among grandchildren and more effective parenting practices among grandparents Rates of grandchild injury are explained by grandparent sex, Kelley et al., 2011 Monserud and Elder, 2011 N’Zi et al., 2016 Nadorff and Patrick, 2018 (continued on next page) Social Science & Medicine 239 (2019) 112476 Participants selected from 80 high schools and 52 middle schools across the U.S.; characteristics of the study population were consistent with a nationally representative sample United States Outcome(s) A.F.A. Sadruddin, et al. Table 2 (continued) Country Study details Sample characteristics Context of care regard to safety and injury prevention 144 grandparents who are past or current caregivers for their grandchildren Participants tended to be of lower 84 children, 9 yr, raised solely by grandparents and enrolled in socioeconomic status and births outside of marriage were common the Fragile Families Child Wellbeing Study birth cohort Outcome categorya Outcome(s) Reference depressive symptoms, and safety knowledge PH; SB; CE BMI, physical health condition; externalizing & internalizing behavior challenges, mental health condition, social skills; academic performance, cognitive ability, developmental delay United States Longitudinal study of how types of grandmother involvement influence grandchildren's functioning United States Cross-sectional analysis of grandparental care and attachment representations in association with child behavior problems United States Cross-sectional analysis of the association between household arrangement and opioid misuse among justice-involved children United States Cross-sectional analysis examining relationship between grand-maternal parenting and grandchild psychological adjustment United States 343 custodial grandmothers and Randomized controlled trial to compare the effectiveness of their grandchildren, 4–12 yr, three intervention models for from four different states reducing children's externalizing and internalizing behavior challenges Children resided with grandparents SB exclusively for at least three months, and families were fluent speakers of English, identifying as Caucasian, African American or Hispanic/Latinx Externalizing & internalizing behavior challenges United States Cross-sectional analysis of the effects of grandmothers' coping resources on their grandchildren's behaviors 733 grandmothers and their grandchildren, 3 mo–16 yr, with the children's parents absent for 3+ months Children resided with grandparents SB exclusively for at least three months, and participating families were diverse with regard to marital Externalizing & internalizing behavior challenges Participants were from low-income SB families, residing in urban neighborhoods in Boston, Chicago and San Diego SB Participants resided in the Midwestern region of the U.S., with about half of families recruited via community support groups, state extension offices, the Department of Health and Family Services kinship care unit, and church groups, and the other half from a medium-minimum security state prison for women SB Living in a single-parent home, 79,960 justice-involved which is common among justicechildren, mean age 14 yr, involved children, is linked to involved with the Florida Department of Juvenile Justice likelihood of opioid misuse, putting youth in these circumstances at higher risk of opioid-related overdose 343 children, 4–12 yr, and their Children resided with grandparents SB exclusively for at least three grandmothers enrolled at months, and families were fluent baseline in an RCT for speakers of English, identifying as improving family well-being Caucasian, African American or Hispanic/Latinx Externalizing & internalizing behavior challenges Attachment, externalizing & internalizing behavior challenges Parenting practices are similar between grandparent families and other structures, but child wellbeing and academic outcomes are poorer, with higher rates of externalizing behavior challenges, grade repetition in school, and delinquent activities among children raise by grandparents Young adolescents with a custodial grandmother present more externalizing behavior challenges over time than their similarly economically disadvantaged peers Children include more relationship violence in their representations of relationships when grandparents report more depressive symptoms; children exhibit fewer externalizing behavior challenges when grandparents are more responsive and when children hold positive family representations Pilkauskas and Dunifon, 2016 Pittman, 2007 Poehlmann et al., 2008 Substance use Justice-involved children are 28% as Shaw et al., 2019 likely to report opioid misuse in the past 30 days as those living in a single-parent home; grandparentheaded homes may be associated with increased access to opioids Externalizing & internalizing behavior challenges Grandmothers' reports of grandchild Smith et al., 2018 externalizing and internalizing behavior challenges are related to grandmother distress and parenting practices, but children's self-reported symptoms are unrelated to grandmother factors Smith et al., 2018b Children exposed to the cognitive behavioral therapy (CBT) and behavioral parent training (BPT) interventions have better outcomes than those exposed to the information-only control program; there is little difference in effectiveness between CBT and BPT for children's behavior Ineffective parenting has a direct Smith et al., 2015 effect on externalizing and internalizing behavior challenges in grandchildren; custodial (continued on next page) Social Science & Medicine 239 (2019) 112476 Cross-sectional analysis of child outcomes and grand-family characteristics, compared with other family structures 17 United States 2402 children, 10–14 yr, and caregivers in low-income, urban neighborhoods from a household-based, stratified random sample survey Children, 3–7.5 yr, with no parent at home; 37 families with a parent in a state prison and 42 families recruited via a community agency Key findings A.F.A. Sadruddin, et al. Table 2 (continued) Social Science & Medicine 239 (2019) 112476 Ziol-Guest and Dunifon, 2014 Smith and Palmieri, 2007 well studies answered the research questions—rather than the precision of outcome findings and publication bias—because our review aimed to improve research decisions, rather than reach clinical guidelines. We measured the inter-rater reliability (Kappa = 0.889) of ratings with a subsample of 100 papers; a kappa score of 0.80 or above reflects good trustworthiness of decisions (Shea et al., 2017). A third author reviewed in full the 206 publications included in the final database, to ensure consistency and adherence to the search strategy (PRISMA, Fig. 1). Characteristics of the study population were consistent with a nationally representative sample 67,588 children, 0–17 yr, from two waves of the National Survey of America's Families Cross-sectional analysis of association between complex living arrangements and children's health United States Our analyses proceeded in two steps. First, we tabulated the total sample (n = 206) of published work with respect to geographic location, study design, sample characteristics, types of care in skippedgeneration or multigenerational settings, and child health and development outcomes (Table 1). We recorded whether studies were longitudinal, cross-sectional, case control, randomized controlled trials, descriptive retrospective, or ethnographic, and whether children were living with a grandparent, with or without their parents. We documented methodological quality, such as sample sizes and the characteristics of population sub-groups under consideration. We categorized outcomes as follows: physical health, socio-emotional and behavioral health, and cognitive and educational development. Where categories overlap (e.g., social, psychological, socio-emotional, and socio-behavioral outcomes), we followed the rubric adopted in recent child development literature (e.g., Tanskanen and Danielsbacka, 2017). Second, we analyzed the review database to evaluate the quality of literature to date and the factors potentially influencing the heterogeneity of grandparent care and child health. We followed a structured approach to evidence synthesis to examine the limitations, relevance, coherence, and adequacy of data (Johnson and Hennessy, 2019; Lewin et al., 2015). At a descriptive level, we identified the types of households featuring grandparent involvement and the range of child outcomes (Table 1). At an interpretive level, we identified major findings, intervening variables, and research gaps for skipped-generation (Table 2) and multigenerational (Table 3) families. We included in Table 2 the total sample of studies pertaining to custodial grandparents caring for children in skipped-generation families, as these publications were small in number (n = 35). To generate comparable data in Table 3, we tabulated one multigenerational study per country or multicountry combination (n = 48), to indicate the global scope of research to date; where there were multiple studies per country, we made a purposive selection based on study design (longitudinal over crosssectional) and largest sample size (capturing more diverse and/or nationally representative samples). In both Tables 2 and 3, we provided additional information on contexts of care, child outcome(s), and key findings. This provided a further opportunity to appraise the body of data, with a view to generate insights from global exemplars in the literature. CE, cognitive and educational development; PH, physical health; SB, socio-emotional and behavioral health. Children resided with grandparents SB exclusively for at least three months 733 grandmothers (NIMH study); 9878 caregivers (2001 National Health Interview Survey) and children, 4–17 yr Cross-sectional analysis of custodial grandchildren's risk of emotional and behavioral problems compared with peers United States 2.3. Data analysis 3. Results 3.1. The global evidence base on types of grandparental care Our search captured 206 publications (Table 1). As expected, this body of work included studies of skipped-generation families, where grandparents were directly responsible for the day-to-day care of the child (denoted C for grandparent custodial care, n = 35); multigenerational families, where grandparents and parents were caregivers together (denoted MG for multigenerational care, n = 154); and both types of care (denoted C & MG, n = 17). We identified some ambiguity in a handful of studies: we included, in the category of custodial care, works specifying that grandparents were primarily responsible for the child while providing no information on residence, and included, in the category of multigenerational care, non-cohabitating grandparents who had regular contact with the child. The distribution of studies was as a Children resided with grandparents SB exclusively for at least three months, and participating families identified as White or Black 733 grandmothers and their grandchildren, 3 mo–16 yr, with the children's parents absent for 3+ months Cross-sectional analysis of an adaptation of the Family Stress Model PH; SB grandmother coping resources have indirect effects (through ineffective parenting) Adjustment Dysfunctional parenting mediates the effect of grandmothers' distress on grandchildren's adjustment, especially with regard to externalizing behaviors Emotional & behavioral challenges Custodial grandchildren fare worse than others across all domains of emotional and behavioral challenges measured by the Strengths and Difficulties Questionnaire Physical health condition; mental Children in custodial grandmother health condition care have better health outcomes, but physical and mental health problems remain high; these children are also more likely to have a limiting condition than many peers status, education, residential locale, work status, and income United States Table 2 (continued) Sample characteristics Context of care Study details Country Outcome categorya Outcome(s) Key findings Reference Smith et al., 2008 A.F.A. Sadruddin, et al. 18 Study details Sample characteristics Context of care Outcome categoryb Outcome(s) Main findings Reference Australia Cross-sectional analysis of the intergenerational transmission of dietary behaviors 114 members of 27 threegeneration families of Anglo-, Chinese- and Italian-Australian descent Grandmothers have much influence on the diets of grandchildren in some cultural contexts PH Eating behaviors Rhodes et al., 2016 Bangladesh Cross-sectional analysis of the effect of maternal mortality and adoption on child survival Grandmothers, more so than other family members, likely ensure adequate care of grandchildren PH Mortality Brazil Longitudinal study of the influence of grandmothers on breastfeeding practices All available records of children born shortly before their mothers' death and children born immediately before the last child born before their mothers' death, from 1974 to 2005 601 mothers who gave birth to healthy babies at a university general hospital in Porto Alegre Regardless of cultural context, grandmothers (and mothers) exert a great deal of influence over family food choice and share a bi-directional relationship with grandchildren that skips the parent generation The probability of survival for children is much higher if they stay in their own household and a grandmother is present Grandmothers may contribute to suboptimal infant feeding practices through advice given to new mothers about the use of other liquids or milks PH Infant feeding practices Susin et al., 2005 Brazil & Peru Cross-sectional analysis of the effects of the presence of a grandparent on children's wellbeing Cross-sectional analysis of the effect of grandparental support on adolescents depression 19,285 individuals living in 3623 households in Brazil, and 19,042 individuals living in 3585 households in Peru 62 adolescents, mean age 14.2 yr, and their parents and grandparents PH; CE Height; academic performance SB Depressive symptoms Adolescents exposed to harsh parenting practices have lower levels of depression when they have supportive grandparents Botcheva and Feldman, 2004 Cameroon Cross-sectional analysis of adolescent sexual behavior 671 youths living in Bemenda Comparison of children living with a grandparent in the household and those without the presence of a grandparent in the home Parents and children lived with grandparents during a prolonged period of economic depression as a result of a transition from a centralized communist economy to one of market-based capitalism; economic stressors led to formation of three-generation households Adolescents living in different household structures ranging from high standard of living to poor households Mothers are less likely to exclusively breastfeed in the first month if grandmothers suggest the introduction of water or tea, and in the first six months if grandmothers suggested other kinds of milk; daily contact with the maternal grandmother is a risk factor for abandoning exclusive breastfeeding Grandparent co-residence is associated with decreased height for age in Peru, but not in Brazil SB Sexual Behavior Rwenge, 2000 Chile Longitudinal study of the effect of household membership transitions on child vocabulary and behavior 15,175 children, 0–5 yr at initial survey and 2 yr older at follow up, from a nationally representative sample CE Language development China Longitudinal study of associations between infant feeding preferences in mothers' social networks and rates of breastfeeding 1172 mother-infant dyads recruited from four public hospitals in Hong Kong Reasons for co-residence with grandparents in Chile include decreasing marriage rates and increasing teen fertility rates; grandparent may provide care as a transition to cohabitation of the mother and father Social networks play an important role in supporting new mothers, and grandparents occupy a space of increased influence Adolescents living with grandparents are significantly less likely to be sexually active and to have multiple partners compared with those living with their parents Children who transition out of grandparent co-residence have worse vocabulary scores, but no significant changes in behavior PH Infant feeding practices Bai et al., 2016 China & United States Cross-sectional analysis of crosscultural differences in grandparent 238 grandparent caregivers in the United States and 106 in China U.S. grandparents assume care of children due to events such as parental death, incarceration, or SB Adjustment Relatively few maternal and paternal grandparents support exclusive breastfeeding, and living with a paternal in-law is associated with risk of early breastfeeding cessation Chinese grandparents report greater grandchild psychological dysfunction than American grandparents, which 19 Country Bulgaria A.F.A. Sadruddin, et al. Table 3 Studies of grandparent care in multi-generation families (n = 48, one exemplara per country or multi-country study). Razzaque et al., 2014 Pérez et al., 2007 Reynolds et al., 2018 (continued on next page) Social Science & Medicine 239 (2019) 112476 Wang et al., 2019 Country Study details Sample characteristics caregiving and grandchild adjustment Context of care Outcome categoryb Outcome(s) divorce, while Chinese grandparents become primary caregivers primarily due to parents' migration for work 20 Cross-sectional analysis of the effects of grandparental economic, cultural, and social capital on child educational attainment 2383 children participating in the Danish Longitudinal Survey of Youth Grandparents' economic, cultural, and social capital may exert unique effects on grandchildren's development and aspirations CE Choice of secondary education Ethiopia Cross-sectional analysis of the effect of kin presence on child mortality and growth Anthropometric and demographic data for 700 children, ≤16 yr, censuses in 1999 and 2003 PH Growth, mortality Europe & Israel Cross-sectional analysis of the effects of grandparental resources on grandchild educational attainment 24,447 individuals, 50+ yr, participating in the Survey of Health, Aging & Retirement in Europe Shifting marriage patterns in contexts of chronic food and water insecurity result in increasing rates of matrilocal residence and access to maternal kin; this increases access to two sets of grandparents providing care In welfare states, grandparental resources may provide a substitute for limited parental resources, buffering educational, financial, and material difficulties CE Educational attainment Finland Cross-sectional analysis of the effects of grandparental education and SES on grandchild completion of secondary education 70,845 individuals, 20 yr, participating in the Finnish Growth Environment Panel Grandparents may influence grandchildren's educational attainment through resources, positive influences of contact, or access to resources in the extended family network CE Educational attainment The Gambia Longitudinal study of the effects of living kin on child survivorship All available records of births and deaths from 1950 to 1974 in western region villages participating in medical surveys Care and resources provided by maternal kin in patrilineal and patrilocal contexts with seasonal food scarcity may provide a buffer that supports child survival PH Mortality Germany Longitudinal study of the effects of living near grandparents on infant survival Genealogies from German villages from 1412 to 1912 Paternal and maternal grandparent investment strategies may differ when both live close to their grandchildren PH Mortality Ghana Cross-sectional analysis of the effect of household structure on timing of adolescents' sexual debut 2387 adolescents, 12–19 yr, enrolled in the 2004 Adolescent Survey Co-residence with grandparents may follow maternal divorce and lack of household wealth SB Sexual behavior Greece Cross-sectional analysis of the effect of demographic factors on Representative sample of 729 children, aged 9–13 yr Grandparents in this context may have experienced hunger and PH Overweight & obesity may be due to greater use of an authoritative parenting style and cultural differences related to the stigma of being raised by a grandparent Grandparents' economic and social capital have no direct effects on grandchildren's choice of secondary education, but grandchildren of those with more cultural capital are more likely to enroll in academically orientated tracks Living with a grandmother has a positive effect on child survival; specific height benefits are observed when children live with maternal grandmothers Higher grandparental education and financial resources are positively associated with grandchild educational attainment, providing some buffering of limited parental education and financial resources Grandparents' educational attainment and SES have limited effects on grandchildren's educational attainment; contact with maternal grandmothers increases grandchildren's educational achievement, particularly in lowincome families, and paternal grandmothers provide access to resources in extended family networks Children without living maternal grandmothers have lower survival rates than those with living maternal grandmothers, but the presence or absence of all other grandparents has no effect Maternal grandparental proximity has no effect on child survival but paternal grandmother proximity exerts protective effects through 12 months, while paternal grandfather proximity exerts a negative effect Adolescents living with grandparents have earlier sexual debut than those living with parents, partially due to parental monitoring and household wealth Having a maternal grandmother as the primary caregiver is associated Reference Møllegaard and Jæger, 2015 Gibson and Mace, 2005 Deindl and Tieben, 2017 Lehti et al., 2019 Sear et al., 2002 Kemkes-Grottenthaler, 2005 Tenkorang and Adjei, 2015 Moschonis et al., 2010 (continued on next page) Social Science & Medicine 239 (2019) 112476 Denmark Main findings A.F.A. Sadruddin, et al. Table 3 (continued) Country Study details Sample characteristics prevalence of childhood overweight and obesity Guatemala Cross-sectional analysis of the effects of grandparent contact and investment on child health 2892 mothers and 3370 children enrolled in the Guatemalan Survey of Family Health Hungary Cross-sectional analysis the effect of grandparental attitudes on physical activity 509 children, 10–14 yr, their parents and grandparents India Cross-sectional analysis of the effect of child care strategy on child health and nutritional status 920 children, 1–3 yr, living in periurban Punjab Indonesia Descriptive retrospective secondary analysis of a quasiexperimental study promoting exclusive breastfeeding Longitudinal study of the effects of maternal employment and child care strategy in infancy on children's non-cognitive skills Cross-sectional analysis of emotional closeness to grandparents and adolescent emotional and behavioral difficulties 163 mother-infant dyads and the infants' grandmothers Ireland 21 Israel 11,314 children, assessed at 9 mo and at 3 yr, whose families participated in the infant cohort of the Growing Up in Ireland study 2751 Arab and Jewish middle- and high-school students, 12–18 yr, grades 7–11 Context of care difficulties finding food as children and may overfeed grandchildren despite changes in food availability Guatemala is one of the poorest countries in the world and grandparents may provide financial assistance and personal support to mothers and children Grandparents may contribute to physical and mental health of grandchildren when parents work extend hours or multiple jobs Working mothers may use different childcare strategies (i.e., grandmothers, siblings, etc.) that have implications for children's health Early breastfeeding rates are below target rates in Indonesia and grandmothers may improve rates through support of mothers Low SES families are more likely to depend on forms of unpaid child care, such as grandparents Outcome categoryb Outcome(s) PH Height for age PH Physical activity PH Height for age, weight for age PH Infant feeding practices SB Emotional & behavioral challenges Grandparents increasingly care for children in contexts with high rates of dual-parent employment; the effects of grandparents on adolescent wellbeing may vary by cultural context SB Adjustment, prosocial behavior 43,046 children born in 2001, followed from 2.5 to 13 yr, with families participating in the Longitudinal Survey of Newborns in the 21st Century Grandparents may exert unique effects on grandchild obesity risk, which may change as children age and spend less time at home PH Overweight & obesity Kenya Cross-sectional analysis of an intervention using dialogue-based groups to promote optimal infant feeding practices Cross-sectional analysis of the association between grandmother co-residence and child school enrollment 509 mothers, fathers, and grandmothers participating in the intervention PH Infant feeding practices CE School enrollment Cross-sectional analysis of proximity to matrilineal kin on child mortality 1635 children from two villages located by Lake Malawi in the southern region of the country Inventions targeting grandparents and fathers may improve optimal breastfeeding rates in a context with high rates of child malnutrition High rates of HIV prevalence result in a high percentage of children living with the extended family; children live with grandmothers in cases of mothers dying or being unaffiliated with the household Women living close to matrilineal grandparents may have increased opportunities for help and other resources PH Mortality Malawi Mexico PH Having a living paternal grandmother is negatively associated with height, while contact with maternal grandparents is weakly associated with height for age Grandparents contribute to the development of healthy attitudes toward physical activity and activity patterns of grandchildren Children of working mothers cared for by grandparents have better outcomes than those in other child care types (siblings and other adults) Lack of grandmother support is associated with shortened early breastfeeding; grandmothers tend to promote introduction of other foods Unpaid grandparent childcare during infancy is associated with higher rates of emotional and behavioral challenges in children at 3 years Emotional closeness to grandparents is associated with fewer emotional symptoms and hyperactivity, and increased rates of prosocial behavior; the effect is greater in Israeli Arab adolescents despite lower levels of emotional closeness to grandparents The effects of grandparent coresidence on risk of overweight and obesity in grandchildren is apparent at age 5.5 years, with highest likelihood of obesity at ages 10 and 13 for boys, and at age 11 for girls Interventions that improve grandmothers' knowledge of infant feeding and encourages social support improve some infant feeding practices Children living with grandmothers and mothers are as likely to be in school; co-residence with a grandmother increases likelihood of school enrollment in orphans and children with absent mothers Presence of maternal grandmothers is associated with higher mortality rates; presence of local paternal grandmothers is weakly protective, compared to paternal grandmothers living in another village Sheppard and Sear, 2016 Lakó, 2014 Gurupdesh et al., 2013 Susiloretni et al., 2015 McDonnell, 2016 Attar-Schwartz and Khoury-Kassabri, 2016 Ikeda et al., 2017 Mukuria et al., 2016 Parker and Short, 2009 Sear, 2008 Schmeer, 2013 (continued on next page) Social Science & Medicine 239 (2019) 112476 Longitudinal study of the effect of household composition on rates of overweight and obesity of children 10,641 children, 6–16 yr, participating in the Lesotho Demographic and Health Survey Reference with higher odds of overweight or obesity Japan Lesotho Main findings A.F.A. Sadruddin, et al. Table 3 (continued) Country Outcome(s) Main findings Iron-deficiency anemia Co-resident grandmothers are associated with reduced rates of child anemia, independent of changes in family structure of parents PH Infant feeding practices CE Educational attainment PH Infant feeding practices SB Social skills Rates of optimal breastfeeding practices and correct timing of introduction of complementary foods and liquids are higher in households with grandmothers with correct knowledge of infant feeding practices Effects of grandparental resources are not significant after controlling for parental characteristics; the effects remain insignificant after including measures of grandparental closeness and involvement Mothers and grandmothers employ different methods of feeding infants and young children, with mothers more likely to use a bottle and grandmothers more likely to hand- or force-feed children; both groups utilize cups, although mothers do this more frequently, and the majority of mothers believe force-feeding can be dangerous; grandmothers note more social support around hand feeding The relationship between parental authoritativeness and children's social competence is moderated by emotional closeness to grandparents SB Externalizing & internalizing behavior challenges, prosocial behaviors, substance use Grandparent involvement is positively associated with prosocial behavior, but not with externalizing and internalizing behavior challenges, or with substance use Profe and Wild, 2017 PH; SB Growth; drinking behavior, smoking behavior Ko, 2019 Co-resident grandmothers and grandfathers who smoke may influence the development of smoking habits of grandchildren SB Smoking behavior As the data are nationally representative across multiple countries, families live in an array of CE School enrollment Grandparents may invest in grandchildren in different ways; grandmothers' pension income is associated with granddaughters' height and lower rates of drinking, while grandfathers' pension income was associated with decreased rates of smoking Living with a grandparent who smokes reduces the likelihood of grandchildren being non-smokers, though the effect is stronger for those who have parents who smoke Grandparents over 60 yr have the greatest impact on grandchildren's schooling and grandmothers are Context of care Longitudinal study of the effect of changes in family structure on rates of child anemia 4649 children, 3–12 yr, participating in the Mexico Family Life Survey Nepal Cross-sectional analysis of the effect of grandmothers' breastfeeding knowledge on mothers' infant feeding practices 1399 infants and toddlers, 16–24 mo, living with grandparents Co-resident grandmothers may have beneficial effects on child nutrition through their expanded social networks or detrimental effects by disrupting mothers' decision making Grandparents knowledge of breastfeeding may influence mothers' feeding practices for their young children The Netherlands Cross-sectional analysis of the effects of grandparental education, occupational status, and cultural resources on grandchildren's educational attainment 1540 children participating in the Longitudinal Internet Studies for Social Sciences survey Nigeria Cross-sectional analysis of the methods used by mothers and grandmothers to feed children under 2 yr and the factors that affect their choice of weaning methods 240 mothers and 246 grandmothers of children under 2 yr in a southwest Nigeria town Pakistan Cross-sectional analysis of the relationship between parenting style and social competence, and the moderating effect of emotional closeness to grandparents Cross-sectional analysis of the effects of parental and grandparental involvement on adolescent mental health and substance use 99 triads of mothers, fathers and adolescent children, 13–18 yr, in joint three-generation family systems South Korea Cross-sectional and longitudinal analyses of the effects of grandparents' pensions on the health of co-resident grandchildren 759 children, in grades 4–6, living with a relative participating in the Korea Welfare Panel Study Spain Cross-sectional analysis of associations between living with a family member that smokes and adolescent smoking 32,234 students, 14–18 yr, that participated in the 2010 State Survey on Drug Use in Secondary School Students Sub-Saharan Africa Cross-sectional analysis of the association between children's 917,788 children, 7–15 yr, in three-generation households South Africa 512 students in grades 8 and 9, mean age 14 yr, from metropolitan Cape Town Grandparents may contribute to the educational attainment of grandchildren through serval types of resources, including their education, occupational status, and cultural resources Igbo-Ora women are typically selfemployed, processing farm produce or trading; they usually carry their infants and young children with them as they work, which can result in extended breastfeeding, but women must also travel to trade and do not always bring their children with them for this part of their work so they may be away from home for days or weeks at a time Grandparents provide emotional support to children living in threegeneration households Apartheid-era policies and social transformations, such as the HIV epidemic and increasing participation of women in the workforce, have resulted in many three-generation households Grandfathers and grandmothers invest resources in co-resident grandchildren, particularly after an old-age pension was introduced in 2007 Outcome categoryb Reference Karmacharya et al., 2017 Bol and Kalmijn, 2016 Titiloye and Brieger, 2009 Akhtar et al., 2017 Escario and Wilkinson, 2015 Schrijner and Smits, 2018b (continued on next page) Social Science & Medicine 239 (2019) 112476 Sample characteristics 22 Study details A.F.A. Sadruddin, et al. Table 3 (continued) Country Study details Sample characteristics schooling and grandparent coresidence Outcome categoryb Outcome(s) settings under widely varying circumstances Academic achievement CE Cognitive development PH Height, weight CE Non-verbal intelligence SB Sexual behavior Children reside with at least one grandparent and at least one parent in the home PH Accidental injury 8250 mothers enrolled in the Early Childhood Longitudinal Study, and 4053 mothers enrolled in the Fragile Families and Child Wellbeing Study The relationship between grandparent co-residence and breastfeeding practices likely varies as a function of the health of grandparents and the socioeconomic status of the family PH Infant feeding practices 1314 caregivers and 591 children, 12–16 yr Extended household structures are common in many low- and middleincome countries, and grandparents may serve as either a risk or protective factor for children The effect of grandparent coresidence may vary in relation to parents' age and marital status, as well as cross-culturally SB Externalizing & internalizing behavior challenges SB Externalizing & internalizing behavior challenges Taiwan Cross-sectional analysis of grandparental co-residence in single-parent households and children's cognitive development Tanzania Cross-sectional analysis of the effects of father absence on food security 3136 children, under 5 yr from 56 villages participating in the Whole Village Project Thailand Cross-sectional analysis of the effect of child rearing by grandparents on child development 320 children, 6–12 yr, from Phrae province Uganda Longitudinal study of the effect of household composition on early sexual debut 1940 adolescent girls, 16–19 yr, participating in the Rakai Community Cohort Study United Kingdom Longitudinal study of the effects of grandparent co-residence on child household injuries 12,319 children, assessed at 9 mo and 3 yr, from the Millennium Cohort Study United States Longitudinal study of grandparent co-residence and breastfeeding initiation and duration Vietnam Cross-sectional analysis of risk factors for poor child mental health in a nationallyrepresentative epidemiological sample Cross-sectional analysis of grandparent co-residence and child mental health 23 Cross-sectional analysis of the effect of grandparental wealth on the educational achievement of grandchildren All available records of children in grade 9 born from 1980 to 1996 with traceable grandparents and whose information was included in national registries 19,531 children in grade 7 participating in the Taiwan Educational Panel Survey 4582 children from six western and eastern European countries participating in the School Children Mental Health in Europe survey Even in egalitarian contexts such as Sweden, wealth inequality within a family may exert strong effects on grandchildren's academic achievement The effects of grandparental coresidence on children's outcomes in single-parent families may depend on whether children live with their mother or father Grandparents may buffer child health in developing contexts with high rates of father absence, due to death or parental separation Children reside with grandparents when parents primarily when parents migrate for work, although a small subset do so when a parent dies or divorces or when co-residing parents have no childcare role Children reside within a variety of household structures, including with both parents or with a grandparent in the household Main findings particularly important for girls—especially those whose mothers died or do not live in the home with them—with regard to school enrollment; however, the impact of grandparents is less when opportunities for schooling are limited across the community Grandparental wealth is associated with children's grade point average in grade 9; this is only partially mediated by parental wealth and socioeconomic characteristics Grandparent co-residence is positively associated with children's cognitive scores; the effect is greater for those living in single-father families Although children living apart from their fathers have lower height-forage z-scores, this effect is attenuated for children living with grandparents Children cared for by grandparents are two times more likely to have delayed development than those with parental care; access to basic needs and services, and family income, are related to development Girls living with grandparents have significantly higher odds of sexual debut before age 16 compared with those living with both biological parents Grandparent co-residence is associated with lower probability of child home injury, with stronger effects for boys Breastfeeding initiation is lower in grandparent co-resident families in the Fragile Families study, but not in the nationally representative sample; co-residence is associated with reduced breastfeeding duration in both samples Grandparent co-residence is neither a risk factor nor is it protective for child mental health Grandparent co-residence is associated with increased risk of child mental health problems in the total sample and in two-parent homes; the Reference Hallsten and Pfeffer, 2017 Chen, 2016 Lawson et al., 2017 Nanthamongkolchai et al., 2009 Pilgrim et al., 2014 Tanskanen and Danielsbacka, 2017 Pilkauskas, 2014a Weiss et al., 2014 Masfety et al., 2019 (continued on next page) Social Science & Medicine 239 (2019) 112476 CE Sweden 6 European Countries Context of care A.F.A. Sadruddin, et al. Table 3 (continued) Social Science & Medicine 239 (2019) 112476 Academic performance CE Economic and cultural context may influence the effect of household structure on children's academic performance Cross-sectional analysis of the effect of father absence and grandparent co-residence on child cognitive and non-cognitive skills Cross-sectional analysis of the effect of household structure on child academic performance 33 OECD Countries 41 OECD Countries Cross-sectional and longitudinal analyses of the associations between obesity and family structure 8 European Countries a Where there were multiple studies per country, the selection of exemplar was based on study design (longitudinal, over cross-sectional) and largest sample size (capturing more diverse and/or nationally representative samples). b CE, cognitive and educational development; PH, physical health; SB, socio-emotional and behavioral health. Chiu and McBride-Chang, 2010 Radl et al., 2017 Grandparent co-residence is associated with adverse cognitive and non-cognitive outcomes in almost all 33 OECD countries Overall, children with no co-resident grandparents have higher reading scores than those who live with a grandparent in the home; this effect is weaker in collectivist societies Locus of control; academic performance SB; CE Formisano et al., 2014 PH Extended household structures are becoming more common in Europe, creating opportunities for adults other than parents to influence the health of children Children co-reside with grandparents in cases of father absence 12,350 children, mean age 7.8 yr, in the cross-sectional survey; 5236 children (a subset of the 12,350), mean age 5.9 yr, in the longitudinal analysis 259,652 children, 15–16 yr, participating the in the 2012 Program for International Student Assessment 193,841 students, 15 yr, participating in the 2000 Program for International Student Assessment BMI, sum of tricipital & sub-scapular skinfold thickness, waist-to-height ratio effect is not significant in singleparent homes Children living with grandparents in the household have significantly higher BMI z-scores and waist-toheight ratios Outcome categoryb Context of care Sample characteristics Study details Country Table 3 (continued) Outcome(s) Main findings Reference A.F.A. Sadruddin, et al. follows: Africa (n = 17), Asia (n = 68), Australia (n = 3), Europe (n = 32), Latin America (n = 12), Middle East (n = 2), and U.S. (n = 60); we also captured multi-country studies (n = 12). Just over two-thirds of studies were cross-sectional (n = 146, 71%); nearly all others were longitudinal (n = 45), and a single study presented both cross-sectional and longitudinal data (n = 1). The remaining few included randomized controlled trials (n = 6), case control studies (n = 4), descriptive retrospective analyses (n = 3), and an ethnographic account (n = 1). With regard to our first research question (how do grandparents influence child health and development outcomes?), we found a heterogeneous body of work, both in terms of the indicators of grandparent involvement and the mechanisms potentially involved. In terms of indicators, studies of "grandparent effects" included those that examined staying in contact with grandchildren (e.g., data on co-residence or frequency of visits), looking after grandchildren (e.g., data on caregiving behaviors such as bathing, feeding, educating, mentoring, and picking up from school), and/or contributing financial support (e.g., data on school fees or household expenses). The granularity of data permitted or constrained different types of analyses. Some quantitative surveys, for example, examined grandparent effects at a gross, population level, in terms of the associations between types of co-residence and child outcomes (e.g., Deleire and Kalil, 2002): they primarily recorded co-residence data—whether or not grandparents were present in multigenerational family structures. Other surveys examined grandparent effects at the family level, providing data on caregiving roles and investment of resources. These studies could focus attention on multiple indicators of grandparental involvement (Thornberry et al., 2006), assess diverse pathways through which grandparents influenced child outcomes (Xie et al., 2018), or examine associations between child outcomes and different types of grandparent caregiving behaviors, such as feeding, playing, and transporting (Liu et al., 2018). Thus, population-level and family-level approaches, drawing on macro- and micro-level data, have different ways of operationalizing types of grandparent involvement and theorizing the pathways through which grandparents exert an influence on their grandchildren. Cohort results often masked strong heterogeneities: the impacts of grandparents differ significantly for advantaged and less-advantaged households (Del Boca et al., 2018). They also reflected the structural context of state policies. For example, one study on European countries and Israel (Deindl and Tieben, 2017) was able to show, at population level, that the magnitude of grandparent effects can differ with respect to the nature of welfare states. Although grandparents buffer the lack of parental resources in economically disadvantaged families, such effects are attenuated in European states that can offer comprehensive welfare provision to disadvantaged families. Rare were the studies of multigenerational households that sought to differentiate between "exposure effects" and "resource effects" of grandparent involvement, or to differentiate between direct effects to grandchildren and "buffer" effects via support to parents. For example, one study (Lehti et al., 2019) sought to differentiate the legacy, stability, and kin keeper effects of grandparents on grandchildren reaching secondary school. Using Finnish census data, this study tested whether grandparents directly transferred resources across generations (legacy), provided resources for the nuclear family in times of need (stability), or kept families together in ways that benefited the children over a shared lifetime. Another study (Tanskanen and Danielsbacka, 2018) cautioned that "grandparent effects" might even be confounded with a "grandchild effect," since a test of causality, using longitudinal data from the British Millennium Cohort Study, did not support the prediction that grandparental investments improved child wellbeing over time. Although parent-grandparent contact frequency and grandparental financial support were associated with improved cognitive and socio-emotional outcomes for grandchildren, such results were based on between-person 24 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. change that demonstrate that investments in caregiver-child relationships and early-life child outcomes strongly correlate with better laterlife outcomes, including children's future health, wealth, education, and life choices (Tanskanen and Danielsbacka, 2018; Conti and Heckman, 2013). associations and were not replicated in models that compared the same individuals over time. It could thus be that grandparent-grandchild effects result from the different characteristics of children, or their households, rather than grandparental involvement, per se. Some studies included families of non-cohabitating grandparents, who provided daily care for their grandchildren. In the U.S. (e.g., Lau et al., 2019), Chinese-American families have strong traditions of multigenerational care, even with separate residences. In Australia (Rogers et al., 2019), non-cohabitating grandparents were acting as the main informal caregivers for grandchildren (specifically with respect to feeding), whereas in Israel (Attar-Schwartz and Huri, 2019), grandparents were offering advice, counsel, and material support to their grandchildren. Three studies that explicitly compared grandparent custodial and multigenerational care were surveys from China (Liu et al., 2018) and the U.S. (Musil et al., 2018; Pittman and Boswell, 2007). For example, the three-city study by Pittman and Boswell (2007) included custodial, non-cohabitating and co-residing grandmothers, to examine child outcomes by residential status and levels of caretaking responsibility, for low-income and minority population groups. 3.2.2. Global studies of grandparent custodial care To provide more descriptive and interpretive insights into the range of studies included in the global evidence base, we now focus attention on custodial grandparents. Table 2 presents the evidence base (n = 35). It shows a skewed global representation: 23 studies were conducted in the U.S. (including 2 comparing samples in the U.S. with those in China and South Africa, respectively), 5 studies were conducted in China, and 1 study per country was undertaken in Australia, Kenya, South Africa, South Korea, Spain, Taiwan, and Thailand. In the U.S., grandparents care for their grandchildren in response to high rates of parental incarceration, mental illness, drug use, and death (Campbell et al., 2006; Edwards, 2006; Kelch-Oliver, 2011). It is important to note, however, that U.S. research to date has predominantly focused on immigrant or minority population groups (e.g., Somalis, Latinx, Hmong, or African Americans) living in poor, urban neighborhoods. Similar findings, with respect for the reasons why grandparents provide custodial care, are reported in Australia, China, Kenya, South Africa, and Thailand, in cases of parental illness, drug use, migration, and death due to HIV/ AIDS (Downie et al., 2010; Jia and Tian, 2010; Nanthamongkolchai et al., 2011; Oburu, 2005; Sharer et al., 2015). Studies reporting on children's physical health (n = 8), both within and outside the U.S., demonstrated that children raised by custodial grandparents presented poorer outcomes than those raised by their parents. The 5 U.S. studies compared outcomes for grandchildren (9-to17 years old) living with grandparents vs. those living with married parents from low-income Caucasian, minority (i.e., African-American) or immigrant (i.e., non-white Hispanic) communities (Campbell et al., 2006; Conway and Li, 2012; Nadorff and Patrick, 2018; Pilkauskas and Dunifon, 2016; Ziol-Guest and Dunifon, 2014). The Thailand study focused on the nutritional status of children (6-to-12 years old) whose parents immigrated from rural to urban areas (Nanthamongkolchai et al., 2011). The China study reported pesticide poisoning for 1-to-14 year olds (Li et al., 2011), while the Taiwan study reported on rates of HIV treatment adherence among adolescents (Xu et al., 2017). Studies reporting on children's socio-emotional health (n = 30) examined outcomes such as emotional wellbeing, loneliness, externalizing and internalizing behavior challenges, depressive symptoms, suicidal tendencies, and anxiety. The results present a mixed picture. Specifically, 4 studies of African-American children (3-to-16 years old), living with grandparents in low-income, urban neighborhoods, recorded higher prevalence rates of juvenile reoffending, anxiety, depression, and adjustment difficulties (Campbell et al., 2006; Edwards, 2009; Kelch-Oliver, 2011; Poehlmann et al., 2008). Similar findings were reported in China, for “left behind” children (8-to-14 years old), who were 2.5 times more likely to suffer from loneliness and 6.4 times more likely to feel very lonely (Jia and Tian, 2010). By contrast, one U.S. study showed no negative outcomes in terms of psychological wellbeing (Hayslip et al., 2014); a Kenyan study showed no adjustment differences between children raised by grandparents and parents (Oburu, 2005); and an Australian study noted that grandchildren living with grandparents had better emotional wellbeing (Downie et al., 2010). All of the studies reporting on cognitive development (n = 5) showed poor child outcomes, irrespective of age, but focused on families living in socioeconomic adversity. For example, U.S. data from the Fragile Families Child Wellbeing Study showed that grandchildren (age 9) who were raised in custodial households repeated school grades more frequently, compared with grandchildren who were raised in nuclear households (Pilkauskas and Dunifon, 2016). Also in the U.S., 7-to-12 year olds under custodial grandparent care were less likely to complete 3.2. The global evidence on links to child outcomes With respect to our second question (what range of child outcomes is reported globally?), we also found substantial heterogeneity in the evidence base. Studies captured in this review more frequently assessed children's physical health (n = 100) and socio-emotional and behavioral health (n = 91), as compared to the smaller body of published work pertaining to children's cognitive and educational development (n = 51); a number of studies reported multiple categories of outcomes. Of course, the nature of the evidence base will reflect ease of measurement (although mortality and morbidity data are routinely available, there are many challenges inherent in gathering good-quality information on child educational or cognitive performance, especially in low-income contexts) as well as policy concerns (e.g., a concern for infant survival and nutrition vs. child or adolescent development). 3.2.1. Randomized controlled trials We looked to randomized controlled trials to draw potential insights regarding the effectiveness of interventions that involve grandparents in caregiving practices. In Brazil, two clinical trials focused on counseling sessions to promote exclusive breastfeeding and reduce pacifier use in the first 6 months of life, showing positive effects among adolescent mothers cohabitating with maternal grandmothers (Giugliani et al., 2019; Nunes et al., 2011). In the U.S., an intervention study evaluated the efficacy of a home-based, videotape intervention to delay the early introduction of infant complementary foods; a home visitor used the videotape to stimulate discussion about infant feeding and mother-grandmother negotiation skills, such that first-time, adolescent, African-American mothers were able to resist grandmothers' pressures to start complementary feeding (Black et al., 2001). One education program was introduced in an obstetrics clinic to teach "Back to Sleep" messages about safe sleep positions to African-American grandmothers and other senior caregivers, as a means to reduce the risk of Sudden Infant Death Syndrome (SIDS); this trial targeted beliefs and preferences about sleep positions, but found no changes in rates of supine sleep (Flick et al., 2001). Two U.S. trials involved custodial grandparents: a small-scale training program to improve attachment and child behavior problems was delivered to grandmothers and greatgrandmothers in a neighborhood library (N’Zi et al., 2016), and a largescale trial compared the relative efficacy of behavioral parent training, cognitive behavioral therapy, and information-only interventions with respect to grandparenting practices and grandchildren's psychological difficulties (Smith et al., 2018). We see that intervention studies largely target specific caregiverchild interactions and/or caregiver beliefs related to infant health, in clinic, home, or community settings. This is consistent with theories of 25 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. with grandparents strengthened grandchildren's communication skills (Cruise and O'Reilly, 2014), language development skills (Reynolds et al., 2018), and overall educational attainment (Deindl and Tieben, 2017; Perry, 2017). For example, a cross-sectional study conducted among 11,914 students in seventh grade in Taiwan reported that the children living in households with long-term grandparental co-residence received the highest test scores, as compared to those who transitioned out of co-resident households (Pong and Chen, 2010). In terms of specific measures, many studies focused on cognitive ability, language development, and/or problem solving skills. By contrast, the studies in sub-Saharan Africa assessed school enrollment only (Parker and Short, 2009; Schrijner and Smits, 2018), reporting that children living with their mothers and grandmothers were more likely to enroll, but offering no data on educational performance. high school and to enroll in and complete college (Monserud and Elder, 2011), while 2-to-4 year olds lagged behind their peers in outcomes associated with academic achievement (Pittman and Boswell, 2007), and 12-to-18-year-old African-American children were more likely to experience challenges in school (Campbell et al., 2006). Whereas studies in the U.S. focused on educational outcomes, the one study in Thailand examined cognitive development, finding that 6-to-12 year olds raised by grandparents in low-income families were more likely to have below-normal development compared with peers raised by their parents (Nanthamongkolchai et al., 2011). 3.2.3. Global studies of multigenerational care By contrast, research focusing on multigenerational care is geographically diverse. Our exemplars (Table 3), selected to illustrate the global scope of the literature, encompass work in Australia, Bangladesh, Brazil, Bulgaria, Cameroon, Chile, China, Denmark, Ethiopia, Finland, The Gambia, Germany, Ghana, Greece, Guatemala, Hungary, India, Indonesia, Ireland, Israel, Japan, Kenya, Lesotho, Malawi, Mexico, Nepal, the Netherlands, Nigeria, Pakistan, South Africa, South Korea, Spain, Sweden, Taiwan, Tanzania, Thailand, Uganda, the U.K. (as well as England specifically), the U.S., Vietnam, and multi-country groupings. Below, we map the global evidence, with a selection of cases from Table 1 and Table 3. The literature shows a considerable range of findings for children's physical health outcomes, which include: birth weight, infant feeding practices and sleeping positions, growth and developmental status, accidental injury, health conditions, mortality, physical activity, child eating behaviors and obesity. Studies conducted in Australia (Rogers et al., 2019), Brazil (Nunes et al., 2011), China (Li and Liu, 2019), Ethiopia (Gibson and Mace, 2005), The Gambia (Sear et al., 2002), and the U.K. (Tanskanen and Danielsbacka, 2017) showed positive effects on grandchildren's survival, height, weight, and protection from physical injury in the home. By contrast, studies in Japan (Sata et al., 2015), Sweden (Lindberg et al., 2016), Thailand (Nanthamongkolchai et al., 2009), and the U.S. (Krueger et al., 2015) showed that grandchildren co-residing with grandparents had higher BMI, delayed physical development, and decreased physical function. Studies in six different African countries reported on children's physical health outcomes: growth, mortality, birth weight, and stunting (Cunningham et al., 2010; Gibson and Mace, 2005; Kerr et al., 2007; Schrijner and Smits, 2018; Sear et al., 2002; Titiloye and Brieger, 2009). Such diverse results, in comparison to the physical health findings reported for children raised by custodial grandparents, may be due to sampling biases, given that the custodial grandparent families are often the most socio-economically vulnerable in a population. The evidence for children's socio-emotional health outcomes in multigenerational families is also mixed (as was the case for children living with custodial grandparents). In countries such as Bulgaria (Botcheva and Feldman, 2004), India (Renny and Jayasankara, 2016), and South Africa (Profe and Wild, 2017), co-residence with grandparents was associated with lower rates of depression, higher emotional intelligence scores, and heightened pro-social behaviors. In Ghana, children living with their grandparents as well as their parents were more prone to engaging in sexual intercourse at an earlier age, due to less stringent monitoring of behavior (Tenkorang and Adjei, 2015). One study in Pakistan noted that grandparental presence in the home moderated the relationship between parental authoritativeness and grandchildren's social competence (Akhtar et al., 2017). It is certainly difficult to draw firm conclusions regarding causal relationships between household composition and child outcomes, from either cross-sectional (Masfety et al., 2019) or longitudinal (Tanskanen and Danielsbacka, 2018) studies, or to know whether associations are due to changes in family stress or to sampling biases, given that decisions to live in three-generation households may be contingent on whether caregivers must take care of relatively easy vs. more challenging children. In terms of cognitive development, studies showed that co-residence 4. Discussion This study provides a global, comprehensive review of the links between grandparental care and grandchildren's health and development. Previous systematic reviews on this topic have been limited in scope: one reported on physical health outcomes in multigenerational families, excluding custodial grandparents (Pulgaron et al., 2016), while three other systematic reviews focused exclusively on breastfeeding (Negin et al., 2016), child dietary outcomes (Young et al., 2018), or long-term cancer risk factors (Chambers et al., 2017). Given the comprehensive scope of this review, we found a heterogeneous body of work—in terms of indicators, mechanisms, and outcomes—characterizing the literature in skipped-generation and multigenerational families. We identify two important issues in the literature: a limited conceptual attention to operationalizing the diverse ways in which grandparents are involved in the lives of their grandchildren, and a limited attention to contextual variables that might influence pathways of care. Both issues contribute to research gaps in the evidence base and to the mixed results we see in the extant literature, with regard to child outcomes. In response to such research gaps, we present a conceptual framework to help operationalize the parameters of grandparental involvement and to explicitly conceptualize the pathways of care (Fig. 2). The left panel focuses attention on types of care. It brings descriptive and analytical attention to grandparent custodial and multigenerational care in skipped-generation and extended families, which characterizes an important heterogeneity in the evidence base. We differentiated three indicators to measure the influence of grandparents: their contact with the grandchildren, their caregiving behaviors, and their support in terms of resources. These indicators are not always explicit in current literature, and need to be unambiguously reported in forthcoming work. The middle panel describes contexts of care. These are intervening variables that potentially influence caregiver roles and impacts—including interpersonal relationships and structural contexts—guiding understanding of pathways of care in terms of who does what, for whom, when, how, and why. In studies to date, the variables pertinent to interpersonal contexts include the caregiving circumstances dictated by age or health (of the child or grandparent), the cultural importance of child-rearing practices (influencing amount, quality and regularity of care), the caregiving relationships (parent relationship status, quality of parenting relationship), and other variables providing opportunities or barriers to grandparental involvement. For their part, variables pertinent to social and structural contexts help examine the salience of variables such as race and class, age and health, and neighborhood and geographical place, as well as situations of conflict, separation, or crisis. This attention to context helps us document the web of caregiving relationships that may involve mothers, fathers, grandparents, friends, neighbors, siblings, other relatives or paid caregivers, in a wider "community of care". This term was coined to encourage the research and policy communities to take seriously those parenting practices that extend beyond mother-child dyads 26 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. Fig. 2. Pathways of care – A conceptual model of the links between grandparent involvement and child outcomes to guide research and policy. regarding ways to reach out to grandparents, leverage their roles in caregiving, and offer a range of services and programs to support them. For example, financial assistance, health, education or housing support, are ways in which policies can better support family-level care (Black et al., 2001; Pilkauskas, 2012; Pilkauskas and Cross, 2018), while specific interventions can be designed to support parenting behavior and to promote family-level resilience (Kelch-Oliver, 2011). Such initiatives help to provide concrete options for integrating grandparents in policy discussions regarding how families care for the next generation. Importantly, evaluation research is needed on specific interventions, regarding the identification of critical points of engagement with caregivers, how we might strategically engage with the wider community of care (parents, grandparents, and other significant actors in children's lives), and how we systematically evaluate their participation and impacts. (Panter-Brick et al., 2014). Importantly, attention to context also helps us to think about the moderators potentially influencing associations with child health. Finally, the right panel directs our attention to outcomes. Here we consider what impacts are being measured and reported, asking for robust, concrete evidence on the extent to which grandparental care influences child health and development. Our conceptual framework thus identifies modalities of care, contexts in which grandparents are involved, and key outcomes for grandchildren. An explicit conceptual framework is needed to generate more specific research hypotheses regarding impacts on health and development, and to carefully describe what variables, including mediators and moderators, will drive variation in outcomes within and between populations. Empirical studies to date have often concluded that the exact pathways by which the influence of grandparents is exerted are still unclear, and this is because of data limitations (Dunifon and KowaleskiJones, 2007, pp.478–479). For example, there may be direct influences due to help with caregiving, discipline, or cognitive stimulation in the home; there may be transfers of resource, changing the amount of income available in the household, entailing changes in child nutrition or education; and there may be indirect advice or interference with parent-child interactions, which may decrease or add to family stress. Such pathways will vary according to contexts—specifically, the age of the child, the health of the grandparent, socioeconomic status, race, and/or crises of separation, incarceration, migration or conflict. We draw attention to the need for more robust data on three indicators of caregiver involvement—contact, behavior, and support—and a careful description of structural and interpersonal contexts in caregiving research. The framework also speaks to the issue of guiding policy decisions, 4.1. Recommendations for research and policy We began this review to map the global evidence base in order to emphasize the need for developing a more robust approach to understanding grandparent care. We conclude with a call for stronger theory, with respect to hypotheses and analysis, because a lack of attention to indicators, contexts and moderators is one of the reasons why the current evidence base lacks rigor. Moving forward, we need to be more attentive to risk of bias coming from studies with small sample sizes, reporting on high-risk populations only, with low-quality methodology and short evaluation time frames. Future research will also need to robustly evaluate caregiving interventions, in terms of reach, cost-effectiveness, scale-up, and sustainability, and in ways that both help 27 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. raising grandchildren, mainly due to the opioid crisis. Key interventions include providing grandparents with information about how to navigate the school system, offering grandparents and grandchildren access to mental health services, and building social and support networks in affected communities (United States Senate, 2018). In the U.K., workschedule and childcare-tax allowances can afford grandparents more flexibility to participate in their grandchildren's social and cognitive development, especially when parents are unable to fulfill caregiving roles (Fergusson et al., 2008; Lussier et al., 2002). In low- and middleincome countries, and in contexts of fragility (e.g., forced migration, extreme poverty), policy initiatives for grandparental care are non-existent, or emergent at best. There is certainly room for improvement with respect to the kind of support provided to grandparent caregivers. children thrive and promote family-level resilience (Britto et al., 2017; Conti and Heckman, 2013; Panter-Brick et al., 2014). Before this can be done, we need to better identify some of the ways in which grandparents can be involved and supported. This will require drawing upon comprehensive frameworks to generate specific hypotheses, at the population level and/or the family level, regarding the ecological, structural, and cultural drivers of care. Our global review does provide some examples of concrete recommendations to improve research design and policy implications. In the U.S., scholars have called for more comparative and longitudinal work that assesses the health and wellbeing of children raised by grandparents and biological parents, among different socioeconomic, ethnic, and age groups, as well as across geographic locales (Areba et al., 2018; Pilkauskas and Dunifon, 2016; Ziol-Guest and Dunifon, 2014). Other scholars have recommended adopting approaches to strengthen networks of support within and outside the home, paying attention to co-parenting and involving other members of the community, including teachers or counselors in schools and places of worship. For example, Kelch-Oliver (2011) called for practitioners working with grandparent-headed families to adopt a “strengths-based approach” as a way of understanding how grandchildren adjust to living with their grandparents (p.403). It also recommended that future programs involve school counselors to emphasize established coping strategies, and the ways grandchildren maintain bonds with their extended family. In another U.S.-based study, Smith et al. (2015) suggested that taking a family-based approach can further strengthen programs and policies that are focused on improving the health and wellbeing of grandparents and grandchildren who are under their care. Outside the U.S., we also see a few examples of recommendations explicitly addressing grandparent-grandchild interactions. One study in Chile advocated that policy makers look for ways to involve grandparents in interventions when grandchildren transition out of their care (Reynolds et al., 2018). In the U.K., the Fatherhood Institute has produced regular reports, toolkits, and policy recommendations for engaging with a wider range of caregivers who provide support to a growing child. Our knowledge of how policies can effectively support grandparents yet remains limited. Current recommendations are useful, but generic. Grandparents remain largely invisible with respect to policy design, implementation, and evaluation of care: they are rarely integrated in the design or implementation of legal frameworks and service delivery, and seldom included in robust program impact evaluations. Policies on caregiving have often reflected, and perpetuated, structural and cultural biases that prevail in the realm of social welfare, income support, housing or school assistance. In terms of engaging with fathers, for example, gender biases have tended to reproduce a mothering, rather than fathering, model of childrearing (Panter-Brick et al., 2014). In terms of engaging with grandparents, similar biases prevail. For example, grandfathers are often presented in a more negative light, as compared to grandmothers (Bates and Taylor, 2013; Buchanan and Rotkirch, 2016; Bullock, 2007). Such biases counter efforts to understand what happens within families in everyday life and what might boost resilience in times of economic crisis, separation, or conflict. Evidence from recent policy reports published by Zero to Three and the World Bank, emphasize that “grandparents are helping to build the brains of … young children” (Kisner et al., 2017, p.15), and that human development issues need to be brought to light in the context of families experiencing economic crisis (Lundberg and Wuermli, 2012, p.8). Initiatives to support grandparents, especially those providing custodial care, include targeted financial assistance, special types of insurance, housing support, and facilitated school enrollment (Baker et al., 2008; Minkler, 1999; Smith and Beitranm, 2003). In the U.S., the need for targeted assistance has been recognized by the federal government: the Supporting Grandparents Raising Grandchildren Act was signed into law in 2018, to create a task force in charge of developing and disseminating information to help the estimated 2.6 million grandparents who are 5. Conclusion Our review identified two main issues leading to the heterogeneity in the evidence base: operationalizing indicators of grandparent involvement and theorizing the mechanisms potentially involved. There is a need for well-informed research and policy decisions to support grandparents and the children under their care. We present a conceptual framework to better measure and theorize pathways of care, with a view to inform research design and policy implications. Acknowledgments Two co-authors received funding support for research time from the Jacobs Foundation, Switzerland, fellowship number 2015 1170 11, which was not involved in the design, conduct, or write-up of our research. We report no conflicts of interest. We thank Onyx Bruner and Jannik Eggerman for their assistance with the Covidence database for this systematic review. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.socscimed.2019.112476. References Akhtar, P., Malik, J., Begeer, S., 2017. The grandparents' influence: parenting styles and social competence among children of joint families. J. Child Fam. Stud. 26, 603–611. Amorim, M., Dunifon, R., Pilkauskas, N., 2017. The magnitude and timing of grandparental coresidence during childhood in the United States. Demogr. Res. 37, 1695–1706. Areba, E.M., Eisenberg, M.E., McMorris, B.J., 2018. Relationships between family structure, adolescent health status and substance use: does ethnicity matter? J. Community Psychol. 46, 44–57. Arenas, E., 2017. Abuelos at home: differential impact on children's education by family structure. Res. Soc. Stratif. Mobil. 52, 36–48. Attar-Schwartz, S., Huri, Y., 2019. Grandparental support and life satisfaction among adolescents in residential care. Child. Youth Serv. Rev. 96, 70–78. Attar-Schwartz, S., Khoury-Kassabri, M., 2016. The moderating role of cultural affiliation in the link between emotional closeness to grandparents and adolescent adjustment difficulties and prosocial behavior. Am. J. Orthopsychiatry 86, 564–572. Augustine, J.M., Raley, R.K., 2013. Multigenerational: households and the school readiness of children born to unmarried mothers. J. Fam. Issues 34, 431–459. Bai, D.L., Fong, D.Y., Lok, K.Y., Tarrant, M., 2016. Relationship between the infant feeding preferences of Chinese mothers' immediate social network and early breastfeeding cessation. J. Hum. Lactation 32, 301–308. Baker, L., Silverstein, M., Putney, N., 2008. Grandparents raising grandchildren in the United States: changing family forms, stagnant policies. J. Soc. Policy Soc. 7, 53–69. Bates, J.S., Taylor, A.C., 2013. Grandfather involvement: contact frequency, participation in activities, and commitment. J. Men's Stud. 21, 305–322. Bentley, G., Mace, R., 2012. Substitute Parents: Biological and Social Perspectives on Alloparenting in Human Societies. Berghahn Books. Bernie, K., 2013. The factors influencing young mothers' infant feeding decisions: the views of healthcare professionals and voluntary workers on the role of the baby's maternal grandmother. Breastfeed. Med. 9, 161–165. Bishai, D., Trevitt, J.L., Zhang, Y., McKenzie, L.B., Leventhal, T., Gielen, A.C., et al., 2008. Risk factors for unintentional injuries in children: are grandparents protective? Pediatrics 122, e980–987. Black, M., Siegel, E., Abel, Y., Bentley, M., 2001. Home and videotape intervention delays 28 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. early complementary feeding among adolescent mothers. Pediatrics 107, 1–8. Black, M.M., Papas, M.A., Hussey, J.M., Hunter, W., Dubowitz, H., Kotch, J.B., et al., 2002. Behavior and development of preschool children born to adolescent mothers: risk and 3-generation households. Pediatrics 109, 573–580. Bledsoe, C., Isiugo-Abanihe, U., 1989. Strategies of child fosterage among Mende grannies in Sierra Leone. In: Lesthaeghe, R.J. (Ed.), Reproduction and Social Organization in Sub-Saharan Africa. University of California Press, Berkeley. Bol, T., Kalmijn, M., 2016. Grandparents' resources and grandchildren's schooling: does grandparental involvement moderate the grandparent effect? Soc. Sci. Res. 55, 155–170. Botcheva, L.B., Feldman, S.S., 2004. Grandparents as family stabilizers during economic hardship in Bulgaria. Int. J. Psychol. 39, 157–168. Brenner, R.A., Simons-Morton, B.G., Bhaskar, B., Das, A., Clemens, J.D., 2001. Prevalence and predictors of immunization among inner-city infants: a birth cohort study. Pediatrics 108, 661–670. Britto, P.R., Lye, S.J., Proulx, K., Yousafzai, A.K., Matthews, S.G., Vaivada, T., et al., 2017. Nurturing care: promoting early childhood development. The Lancet 389, 91–102. Brook, J.S., Whiteman, M., Zheng, L., 2002. Intergenerational transmission of risks for problem behavior. J. Abnorm. Child Psychol. 30, 65–76. Bruederle, A., Delany-Moretlwe, S., Mmari, K., Brahmbhatt, H., 2019. Social support and its effects on adolescent sexual risk taking: a look at vulnerable populations in Baltimore and Johannesburg. J. Adolesc. Health 64, 56–62. Buchanan, A., Rotkirch, A., 2016. Grandfathers: Global Perspectives. Palgrave Macmillan. Bullock, K., 2007. Grandfathers raising grandchildren. J. Health Soc. Policy 22, 181–197. Campbell, L., Hu, J., Oberle, S., 2006. Factors associated with future offending: comparing youth in grandparent-headed homes with those in parent-headed homes. Arch. Psychiatr. Nurs. 20, 258–267. Chambers, S., Rowa-Dewar, N., Radley, A., Dobbie, F., 2017. A systematic review of grandparents' influence on grandchildren's cancer risk factors. PLoS One 12, 1–28. Chan, K.L., Chen, M., Chen, Q., Ip, P., 2017. Can family structure and social support reduce the impact of child victimization on health-related quality of life? Child Abuse Negl. 72, 66–74. Chaudhry, A.G., Ahmed, A., Hussain, S., Ali, N., Batool, A., 2014. An anthropological study of breastfeeding practices and prevalence of supported liquids among childbearing women in rural Gujrat. Sci. Int. 26, 483–487. Chen, W.-C., 2016. The role of grandparents in single-parent families in Taiwan. Marriage Fam. Rev. 52, 41–63. Chiang, Y.L., Park, H., 2015. Do grandparents matter? A multigenerational perspective on educational attainment in Taiwan. Soc. Sci. Res. 51, 163–173. Chiu, M.M., 2007. Families, economies, cultures, and science achievement in 41 countries: country-, school-, and student-level analyses. J. Fam. Psychol. 21, 510–519. Chiu, M.M., McBride-Chang, C., 2010. Family and reading in 41 countries: differences across cultures and students. Sci. Stud. Read. 14, 514–543. Cisco, J., 2017. Who supports breastfeeding mothers?: an investigation of kin investment in the United States. Hum. Nat. 28, 231–253. Conti, G., Heckman, J.J., 2013. The developmental approach to child and adult health. Pediatrics 131, 133–141. Conway, K.S., Li, M., 2012. Family structure and child outcomes: a high definition, wide angle “snapshot”. Rev. Econ. Househ. 10, 345–374. Courtin, E., Avendano, M., 2016. Under one roof: the effect of co-residing with adult children on depression in later life. Soc. Sci. Med. 168, 140–149. Crosnoe, R., Elder Jr., G.H., 2002. Life course transitions, the generational stake, and grandparent-grandchild relationships. J. Marriage Fam. 64, 1089–1096. Cruise, S., O'Reilly, D., 2014. The influence of parents, older siblings, and non-parental care on infant development at nine months of age. Infant Behav. Dev. 37, 546–555. Cunningham, S., Elo, I., Herbst, K., Hosegood, V., 2010. Prenatal development in rural South Africa: relationship between birth weight and access to fathers and grandparents. Popul. Stud. 64, 229–246. de Brauw, A., Mu, R., 2011. Migration and the overweight and underweight status of children in rural China. Food Policy 36, 88–100. Deindl, C., Tieben, N., 2017. Resources of grandparents: educational outcomes across three generations in Europe and Israel. J. Marriage Fam. 79, 769–783. Del Boca, D., Piazzalunga, D., Pronzato, C., 2018. The role of grandparenting in early childcare and child outcomes. Rev. Econ. Househ. 16, 477–512. Deleire, T., Kalil, A., 2002. Good things come in threes: single-parent multigenerational family structure and adolescent adjustment. Demography 39, 393–413. Dong, H., Manfredini, M., Kurosu, S., Yang, W., Lee, J.Z., 2017. Kin and birth order effects on male child mortality: three East Asian populations, 1716-1945. Evol. Hum. Behav. 38, 208–216. Downie, J., Hay, D., Horner, B., Wichman, H., Hislop, A., 2010. Children living with their grandparents: resilience and wellbeing. Int. J. Soc. Welf. 19, 8–22. Duarte, R., Escario, J.-J., Molina, J.-A., 2016. Smoking transmission in-home across three generations. J. Subst. Use 21, 268–272. Dunifon, R., Bajracharya, A., 2012. The role of grandparents in the lives of youth. J. Fam. Issues 33, 1168–1194. Dunifon, R., Kowaleski-Jones, L., 2007. The influence of grandparents in single-mother families. J. Marriage Fam. 69, 465–481. Dunifon, R.E., Ziol-Guest, K.M., Kopko, K., 2014. Grandparent coresidence and family well-being: implications for research and policy. Ann. Am. Acad. Pol. Soc. Sci. 654, 110–126. Dunifon, R., Near, C., Ziol-Guest, K., 2018. Backup parents, playmates, friends: grandparents' time with grandchildren. J. Marriage Fam. 80, 752–767. Edwards, O., 2006. Teachers' perceptions of the emotional and behavioral functioning of children raised by grandparents. Pyschol. Sch. 43, 565–572. Edwards, O., 2009. Empirical investigation of the psychosocial functioning of children raised by grandparents. J. Appl. Psychol. 25, 128–145. Eitle, D., Niedrist, F., Eitle, T.M., 2014. Gender, race, and delinquent behavior: an extension of power-control theory to American Indian adolescents. Deviant Behav. 35, 1023–1042. Eli, K., Howell, K., Fisher, P., Nowicka, P., 2016. A question of balance: explaining differences between parental and grandparental perspectives on preschoolers' feeding and physical activity. Soc. Sci. Med. 154, 28–35. Emmott, E.H., Mace, R., 2015. Practical support from fathers and grandmothers is associated with lower levels of breastfeeding in the UK Millennium cohort study. PLoS One 10, e0133547. Escario, J.J., Wilkinson, A.V., 2015. The intergenerational transmission of smoking across three cohabitant generations: a count data approach. J. Community Health 40, 912–919. Fan, X.-h., Fang, X.-y., 2010. A comparison of problem behaviors between parent-present children and parent-absent children in different care-taking modes. Chin. J. Clin. Psychol. 18, 232–234. Farrow, C., 2014. A comparison between the feeding practices of parents and grandparents. Eat. Behav. 15, 339–342. Fergusson, E., Maughan, B., Golding, J., 2008. Which children receive grandparental care and what effect does it have? J. Child Psychol. Psychiatry 49, 161–169. Ferreira, T.D.M., Piccioni, L.D., Queiroz, P.H.B., Silva, E.M., Vale, I.N.D., 2018. Influence of grandmothers on exclusive breastfeeding: cross-sectional study. Einstein (São Paulo) 16 (4), 1–7. Flick, L., Vemulapalli, C., Stulac, B., Kemp, J., 2001. Home and videotape intervention delays early complementary feeding among adolescent mothers. Arch. Pediatr. Adolesc. Med. 155, 1231–1237. Flouri, E., Buchanan, A., Tan, J.P., Griggs, J., Attar-Schwartz, S., 2010. Adverse life events, area socio-economic disadvantage, and adolescent psychopathology: the role of closeness to grandparents in moderating the effect of contextual stress. Stress 13, 402–412. Fomby, P., James-Hawkins, L., Mollborn, S., 2015. Family resources in two generations and school readiness among children of teen parents. Popul. Res. Policy Rev. 34, 733–759. Formisano, A., Hunsberger, M., Bammann, K., Vanaelst, B., Molnar, D., Moreno, L.A., et al., 2014. Family structure and childhood obesity: results of the IDEFICS project. Public Health Nutr. 17, 2307–2315. Franca, M.C., Giugliani, E.R., Oliveira, L.D., Weigert, E.M., Santo, L.C., Kohler, C.V., et al., 2008. [Bottle feeding during the first month of life: determinants and effect on breastfeeding technique]. Rev. Saude Publica 42, 607–614. Fuentes, M.J., Bernedo, I.M., 2009. Adaptation and family relationships from the point of view of adolescents under the care of their grandparents. Rev. Psicol. Soc. 24, 53–64. Fuller-Thomson, E., Minkler, M., 2001. American grandparents providing extensive child care to their grandchildren: prevalence and profile. Gerontol. 41, 201–209. Gibson, M., Mace, R., 2005. Helpful grandmothers in rural Ethiopia: a study of the effect of kin on child survival and growth. Evol. Hum. Behav. 26, 469–482. Giugliani, E.R., do Espirito Santo, L.C., de Oliveira, L.D., Aerts, D., 2008. Intake of water, herbal teas and non-breast milks during the first month of life: associated factors and impact on breastfeeding duration. Early Hum. Dev. 84, 305–310. Giugliani, E.R.J., Nunes, L.M., Issler, R.M.S., Santo, L.C.d.E., Oliveira, L.D.d., 2019. Involvement of maternal grandmother and teenage mother in intervention to reduce pacifier use: a randomized clinical trial. J. Pediatr. 95, 166–172. Goh Esther, C.L., 2013. You must finish your dinner. Br. Food J. 115, 365–376. Goodman, C.C., 2012. Caregiving grandmothers and their grandchildren: well-being nine years later. Child. Youth Serv. Rev. 34, 648–654. Gorman, B., Braveman, J., 2008. Family structure differences in health care utilization among U.S. children. Soc. Sci. Med. 67, 1766–1775. Goulette, N.W., Evans, S.Z., King, D., 2016. Exploring the behavior of juveniles and young adults raised by custodial grandmothers. Child. Youth Serv. Rev. 70, 349–356. Grau, J.M., Duran, P.A., Castellanos, P., Smith, E.N., Silberman, S.G., Wood, L.E., 2015. Developmental outcomes of toddlers of young Latina mothers: cultural, family, and parenting factors. Infant Behav. Dev. 41, 113–126. Griggs, J., Tan, J.-P., Buchanan, A., Attar-Schwartz, S., Flouri, E., 2010. ‘They’ve always been there for me’: grandparental involvement and child well-being. Child. Soc. 24, 200–214. Guo, X.-l., 2014. The influence of grandparenting on cognitive development of children: a longitudinal study. Chin. J. Clin. Psychol. 22, 1072–1076. Gurupdesh, K., Parminder, K., Jaswal, S., 2013. Impact of child care strategies on health and nutritional status of children in peri-urban Punjab. Asian J. Home Sci. 8, 539–543. Hallsten, M., Pfeffer, F.T., 2017. Grand advantage: family wealth and grandchildren's educational achievement in Sweden. Am. Sociol. Rev. 82, 328–360. Hamilton, H.A., 2005. Extended families and adolescent well-being. J. Adolesc. Health 36, 260–266. Hank, K., Buber, I., 2009. Grandparents caring for their grandchildren: findings from the 2004 survey of health, ageing, and retirement in Europe. J. Fam. Issues 30, 53–73. Hansen, K., Hawkes, D., 2009. Early childcare and child development. J. Soc. Policy 38, 211–239. Harper, C.C., McLanahan, S.S., 2004. Father absence and youth incarceration. J. Res. Adolesc. 14, 369–397. Hawkes, K., O’Connell, J.F., Blurton Jones, N.G., Alvarez, H., Charnov, E.L., 2017. The grandmother hypothesis and human evolution. In: Chagnon, N. (Ed.), Adaptation and Human Behavior: An Anthropological Perspective. Routledge, New York. Hayslip, B., Smith, G. (Eds.), 2012. Resilient Grandparent Caregivers: A Strengths-Based Perspective. Routledge, Taylor & Francis Group, New York, NY. Hayslip, B., Heidemaire, B., Garner, A., 2014. Health and grandparent−grandchild wellbeing: one-year longitudinal findings for custodial grandfamilies. J. Aging Health 26, 559–582. 29 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. Hayslip, B., Fruhauf, C., Dolbin-MacNab, M., 2017. Grandparents raising grandchildren: what have we learned over the past decade? Gerontol. 00, 1–12. He, Q., Li, X., Wang, R., 2018. Childhood obesity in China: does grandparents' coresidence matter? Econ. Hum. Biol. 29, 56–63. Heath, K.M., 2003. The effects of kin propinquity on infant mortality. Soc. Biol. 50, 270–280. Hon, K.L., Ma, K.C., Wong, Y., Leung, T.F., Fok, T.F., 2005. A survey of traditional Chinese medicine use in children with atopic dermatitis attending a paediatric dermatology clinic. J. Dermatol. Treat. 16, 154–157. Ice, G.H., Sadruddin, A.F.A., Vagedes, A., Yogo, J., Juma, E., 2012. Stress associated with caregiving: an examination of the stress process model among Kenyan Luo elders. Soc. Sci. Med. 74, 2020–2027. Ikeda, N., Fuse, K., Nishi, N., 2017. Changes in the effects of living with no siblings or living with grandparents on overweight and obesity in children: results from a national cohort study in Japan. PLoS One 12, e0175726. Ingersoll-Dayton, B., Punpuing, S., Tangchonlatip, K., Yakas, L., 2018. Pathways to grandparents' provision of care in skipped-generation households in Thailand. Ageing Soc. 38, 1429–1452. Ingram, J., Johnson, D., Hamid, N., 2003. South Asian grandmothers' influence on breast feeding in Bristol. Midwifery 19, 318–327. Ishizawa, H., 2004. Minority language use among grandchildren in multigenerational households. Sociol. Perspect. 47, 465–483. Jahromi, L.B., Zeiders, K.H., Updegraff, K.A., Umana-Taylor, A.J., Bayless, S.D., 2018. Coparenting conflict and academic readiness in children of teen mothers: effortful control as a mediator. Fam. Process 57, 462–476. Jamison, C.S., Cornell, L.L., Jamison, P.L., Nakazato, H., 2002. Are all grandmothers equal? A review and a preliminary test of the “grandmother hypothesis” in Tokugawa, Japan. Am. J. Phys. Anthropol. 119, 67–76. Jessel, J., Kenner, C., Gregory, E., Ruby, M., Arju, T., 2011. Different spaces: learning and literacy with children and their grandparents in East London homes. Linguist. Educ. 22, 37–50. Jia, Z., Tian, W., 2010. Loneliness of left‐behind children: a cross‐sectional survey in a sample of rural China. Child Care Health Dev. 36, 812–817. Jia, S.M., Wang, L., Shi, Y.J., Li, P., 2011. [Relations between aggressive behavior and family factors among preschool children in Shanghai]. Chin. J. Epidemiol. 32, 1216–1220. Jia, S., Wang, L., Shi, Y., Li, P., 2016. Family risk factors associated with aggressive behavior in Chinese preschool children. J. Pediatr. Nurs. 31, e367–e374. Jingxiong, J., Rosenqvist, U., Wang, H., Greiner, T., Lian, G., Sarkadi, A., 2007. Influence of grandparents on eating behaviors of young children in Chinese three-generation families. Appetite 48, 377–383. Johnson, B.T., Hennessy, E.A., 2019. Systematic reviews and meta-analyses in the health sciences: best practice methods for research syntheses. Soc. Sci. Med. 233, 237–251. Kamo, Y., 2000. Racial and ethnic differences in extended family households. Sociol. Perspect. 43, 211–229. Kaneko, A., Kaneita, Y., Yokoyama, E., Miyake, T., Harano, S., Suzuki, K., et al., 2006. Factors associated with exclusive breast-feeding in Japan: for activities to support child-rearing with breast-feeding. J. Epidemiol. 16, 57–63. Karmacharya, C., Cunningham, K., Choufani, J., Kadiyala, S., 2017. Grandmothers' knowledge positively influences maternal knowledge and infant and young child feeding practices. Public Health Nutr. 20, 2114–2123. Kasedde, S., Aoide, M., Ross, D., 2014. They are not always a burden: older people and child fostering in Uganda during the HIV epidemic. Soc. Sci. Med. 113, 161–168. Kelch-Oliver, K., 2011. African American grandchildren raised in grandparent-headed families: an exploratory study. Fam. J. 19, 396–406. Kelley, S., Whitley, D., Campos, P., 2011. Behavior problems in children raised by grandmothers: the role of caregiver distress, family resources, and the home environment. Child. Youth Serv. Rev. 33, 2138–2145. Kemkes-Grottenthaler, A., 2005. Of grandmothers, grandfathers and wicked step-grandparents. Differential impact of paternal grandparents on grandoffspring survival. Hist. Soc. Res./Hist. Sozialforsch. 30, 219–239. Kerr, R., Berti, P., Chirwa, M., 2007. Breastfeeding and mixed feeding practices in Malawi: timing, reasons, decision makers, and child health consequences. Food Nutr. Bull. 28, 90–99. Kisner, K., Parlakian, R., MacLaughlin, S., 2017. Grandparents Who Care. Zero to Three, Washington, DC. Knodel, J., Wassana, I.-E., 2004. The economic consequences for parents of losing an adult child to AIDS: evidence from Thailand. Soc. Sci. Med. 59, 987–1001. Ko, H., 2019. Intergenerational effects of old-age pension on children's health and wellbeing. Int. J. Health Plan. Manag. 34, e1208–e1214. Kobayashi, S., Asakura, K., Suga, H., Sasaki, S., 2015. Cohabitational effect of grandparents on dietary intake among young Japanese women and their mothers living together. A multicenter cross-sectional study. Appetite 91, 287–297. Komonpaisarn, T., Loichinger, E., 2019. Providing regular care for grandchildren in Thailand: an analysis of the impact on grandparents' health. Soc. Sci. Med. 229, 117–125. Krueger, P., Jutte, D., Franzini, L., Elo, I., Hayward, M., 2015. Family structure and multiple domains of child well-being in the United States: a cross sectional study. Popul. Health Metrics 13 (6), 1–11. Lakó, J.H., 2014. The issues of the relationship of grandparents and grandchildren in the light of physical activity. Eur. J. Ment. Health 9, 178–194. Lau, J.D., Au, L.Y., Chao, E., Elbaar, L., Tse, R., 2019. The association of grandparent care with childhood overweight and obesity in Chinese American families. Child. Obes. 15, 14–20. Lawson, D.W., Schaffnit, S.B., Hassan, A., Ngadaya, E., Ngowi, B., Mfinanga, S.G.M., et al., 2017. Father absence but not fosterage predicts food insecurity, relative poverty, and poor child health in northern Tanzania. Am. J. Hum. Biol. 29, e22938. Lee, R., Chun, J., Chung, I.-J., Kang, H., Nho, C.R., Woo, S., 2017. Kinship foster care and school adjustment: evidence from a nationally representative sample of children in out-of-home care in South Korea. Child Youth Care Forum 46, 335–356. Lehti, H., Erola, J., Tanskanen, A.O., 2019. Tying the extended family knot—grandparents’ influence on educational achievement. Eur. Sociol. Rev. 35, 29–48. Levetan, J.L., Wild, L.G., 2016. The implications of maternal grandmother coresidence and involvement for adolescent adjustment in South Africa. Int. J. Psychol. 51, 356–365. Lewin, S., Glenton, C., Munthe-Kaas, H., Carlsen, B., Colvin, C.J., Gülmezoglu, M., et al., 2015. Using qualitative evidence in decision making for health and social interventions: an approach to assess confidence in findings from qualitative evidence syntheses (GRADE-CERQual). PLoS Med. 12, e1001895. Li, X., Liu, Y., 2019. Parent-grandparent coparenting relationship, maternal parenting self-efficacy, and young children's social competence in Chinese urban families. J. Child Fam. Stud. 28, 1145–1153. Li, Y., Mora, R., 2016. Re-assessing the impact of the grandparent's income on the infant mortality rate: an evaluation of the old age allowance program in Nepal. World Dev. 87, 333–348. Li, H., Yang, L., Feng, Q.M., Li, C.L., 2011. [Analysis of the risk factors for pesticide poisoning among children in countryside of Guigang city]. Chin. J. Prev. Med. 45, 1103–1107. Li, B., Adab, P., Cheng, K.K., 2014. Family and neighborhood correlates of overweight and obesogenic behaviors among Chinese children. Int. J. Behav. Med. 21, 700–709. Li, B., Adab, P., Cheng, K.K., 2015. The role of grandparents in childhood obesity in China - evidence from a mixed methods study. Int. J. Behav. Nutr. Phys. Act. 12, 91. Li, T., Lam, C.B., Chan, K.K.-S., 2018. Grandparental involvement and young adults’ cognitive and social adjustment: the moderating role of filial piety in Hong Kong. J. Soc. Pers. Relatsh. 35, 999–1018. Li, Q., Liang, F., Liang, W., Zhang, J., Niu, M., Han, Y., 2017a. The influence of different caregivers on infant growth and development in China. Front. Pediatr. 5, 243. Lin, W.-H., Yi, C.-C., 2018. Subjective well-being and family structure during early adolescence: a prospective study. J. Early Adolesc. 39, 426–452. Lindberg, L., Ek, A., Nyman, J., Marcus, C., Ulijaszek, S., Nowicka, P., 2016. Low grandparental social support combined with low parental socioeconomic status is closely associated with obesity in preschool‐aged children: a pilot study. Pediatr. Obes. 11, 313–316. Liu, R.X., 2016. School bonding, peer associations, and self-views: the influences of gender and grandparent attachment on adolescents in mainland China. Youth Soc. 48, 451–469. Liu, L., Fan, L., Hou, X.-Y., Wu, C.-A., Yin, X.-N., Wen, G.-M., et al., 2018. Family childcare types and conduct problem behaviors in young children: the mediation role of caregiver-child interaction. Front. Pediatr. 6, 217. Liu, B., Qing, P., Xiao, S., Liao, F., 2019. The influence of grandparents' indulgence on the health status of the left-behind children in rural areas from the perspective of food consumption: a case study from Hubei Province. Chin. Rural Econ. 1, 1–12. Lu, Y., Zhu, P., Gao, R., Liang, Z.Z., Huang, W., Tao, F.B., 2012. [Association of rearing patterns with diet and temperament traits among infants in urban areas.]. Chin. J. Contemp. Pediatr. 14, 48–53. Lundberg, M., Wuermli, A., 2012. Children and Youth in Crisis: Protecting and Promoting Human Development in Times of Economic Shocks. The World Bank, Washington, DC. Lussier, G., Deater-Deckard, K., Dunn, J., Davies, L., 2002. Support across two generations. J. Fam. Psychol. 16. Ma, W.J., Nie, S.P., Xu, H.F., Xu, Y.J., Zhang, Y.R., 2010. Socioeconomic status and the occurrence of non-fatal child pedestrian injury: results from a cross-sectional survey. Saf. Sci. 48, 823–828. Masfety, V.K., Aarnink, C., Otten, R., Bitfoi, A., Mihova, Z., Lesinskiene, S., et al., 2019. Three-generation households and child mental health in European countries. Soc. Psychiatry Psychiatr. Epidemiol. 54, 427–436. Mazzucato, V., Cebotari, V., Veale, A., White, A., Vivet, J., 2015. International parental migration and the psychological well-being of children in Ghana, Nigeria, and Angola. Soc. Sci. Med. 132, 215–224. McDonnell, T., 2016. Non-cognitive development in early childhood: the influence of maternal employment and the mediating role of childcare. Econ. Soc. Rev. 47 (4) Winter (2016): GROWING UP IN IRELAND - Papers Presented at The Annual GUI Conference, Dublin Castle, 3 December 2015. McFee, R.B., Caraccio, T.R., 2006. “Hang up your pocketbook” – an easy intervention for the granny syndrome: grandparents as a risk factor in unintentional pediatric exposures to pharmaceuticals. J. Am. Osteopath. Assoc. 106, 405–411. Mehta, K., Thang, L., 2012. Experiencing Grandparenthood: An Asian Perspective Singapore. Springer. Metbulut, A.P., Ozmert, E.N., Teksam, O., Yurdakok, K., 2018. A comparison between the feeding practices of parents and grandparents. Eur. J. Pediatr. 177, 1785–1794. Minkler, M., 1999. Intergenerational households headed by grandparents: contexts, realities, and implications for policy. J. Aging Stud. 13, 199–218. Minkler, M., Fuller-Thomson, E., 2005. African American grandparents raising grandchildren: a national study using the census 2000 American community survey. J. Gerontol.: Ser. B 60, S82–S92. Modin, B., Fritzell, J., 2009. The long arm of the family: are parental and grandparental earnings related to young men's body mass index and cognitive ability? Int. J. Epidemiol. 38, 733–744. Modin, B., Erikson, R., Vågerö, D., 2012. Intergenerational continuity in school performance: do grandparents matter? Eur. Sociol. Rev. 29, 858–870. Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., Group, T.P., 2009. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 30 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. 6, 1–6. Mollborn, S., Fomby, P., Dennis, J.A., 2011. Who matters for children’s early development? Race/ethnicity and extended household structures in the United States. Child Indic. Res. 4, 389–411. Møllegaard, S., Jæger, M.M., 2015. The effect of grandparents' economic, cultural, and social capital on grandchildren's educational success. Res. Soc. Stratif. Mobil. 42, 11–19. Monserud, M., Elder, J.G., 2011. Household structure and children's educational attainment: a perspective on coresidence with grandparents. J. Marriage Fam. 73, 981–1000. Moon, R.Y., Omron, R., 2002. Determinants of infant sleep position in an urban population. Clin. Pediatr. (Phila) 41, 569–573. Moon, R.Y., Oden, R.P., Grady, K.C., 2004. Back to Sleep: an educational intervention with women, infants, and children program clients. Pediatrics 113, 542–547. Morioka, H., Itani, O., Jike, M., Nakagome, S., Otsuka, Y., Ohida, T., 2018. Risk factors at birth predictive of subsequent injury among Japanese preschool children: a nationwide 5-year cohort study. J. Dev. Behav. Pediatr. 39, 424–433. Morita, A., Ochi, M., Isumi, A., Fujiwara, T., 2019. Association between grandparent coresidence and weight change among first-grade Japanese children. Pediatr. Obes. 14, e12524. Moschonis, G., Tanagra, S., Vandorou, A., Kyriakou, A.E., Dede, V., Siatitsa, P.E., et al., 2010. Social, economic and demographic correlates of overweight and obesity in primary-school children: preliminary data from the Healthy Growth Study. Public Health Nutr. 13, 1693–1700. Mukuria, A.G., Martin, S.L., Egondi, T., Bingham, A., Thuita, F.M., 2016. Role of social support in improving infant feeding practices in western Kenya: a quasi-experimental study. Glob. Health Sci. Pract. 4, 55–72. Musil, C.M., Rice, H.M., Singer, M., Givens, S.E., Warner, C.B., Zauszniewski, J.A., et al., 2018. Grandchildren's depressive symptoms and perceptions of family functioning: protective and influencing factors. West. J. Nurs. Res. 40, 1319–1338. Nadorff, D., Patrick, J., 2018. Predicting child safety: the effect of custodial grandparents' depressive symptoms, home safety, knowledge, and gender. J. Intergener. Relat. 16, 225–242. Nanthamongkolchai, S., Munsawaengsub, C., Nanthamongkolchai, C., 2009. Influence of child rearing by grandparent on the development of children aged six to twelve years. J. Med. Assoc. Thail. 92 (92), 430–434. Nanthamongkolchai, S., Munsawaengsub, C., Nanthamongkolchai, C., 2011. Comparison of the health status of children aged between 6 and 12 years reared by grandparents and parents. Asia Pac. J. Public Health 2, 766–773. Negin, J., Coffman, J., Vizintin, P., Raynes-Greenow, C., 2016. The influence of grandmothers on breastfeeding rates: a systematic review. Pregnancy Childbirth 16, 1–10. Noah, A.J., 2018. Understanding the relationships between diverse family structures and the development of emotion regulation of Mexican-origin children: population-based estimates. Child Indic. Res. 11, 1515–1530. Notermans, C., 2003. Sharing home, food, and bed: paths of grandmotherhood in East Cameroon. Africa 74, 6–27. Nunes, L.M., Giugliani, E.R.J., do Espírito Santo, L.C., de Oliveira, L.D., 2011. Reduction of unnecessary intake of water and herbal teas on breast-fed infants: a randomized clinical trial with adolescent mothers and grandmothers. J. Adolesc. Health 49, 258–264. N’Zi, A.M., Stevens, M.L., Eyberg, S.M., 2016. Child directed interaction training for young children in kinship care: a pilot study. Child Abuse Negl. 55, 81–91. Oburu, P., 2005. Caregiving stress and adjustment problems of Kenyan orphans raised by grandmothers. Infant Child Dev. 14, 199–210. Odom, E.C., Li, R., Scanlon, K.S., Perrine, C.G., Grummer-Strawn, L., 2014. Association of family and health care provider opinion on infant feeding with mother's breastfeeding decision. J. Acad. Nutr. Diet. 114, 1203–1207. Panter-Brick, C., Burgess, A., Eggerman, M., McAllister, F., Pruett, K., Leckman, J.F., 2014. Practitioner review: engaging fathers – recommendations for a game change in parenting interventions based on a systematic review of the global evidence. J. Child Psychol. Psychiatry 55, 1187–1212. Parker, E., Short, S., 2009. Grandmother co-residence and school enrollment in SubSaharan Africa. J. Fam. Issues 30, 813–836. Pérez, E.R., Turra, C.M., Queiroz, B.L., 2007. Grandparents and grandchildren living together, a better life for the youngest? The cases of Brazil and Peru. Papeles Población 13, 47–75. Pérez-Escamilla, R., Cavallera, V., Tomlinson, M., Dua, T., 2018. Scaling up integrated early childhood development programs: lessons from four countries. Child Care Health Dev. 44, 50–61. Perry, G., 2017. Alloparental care and assistance in a normatively patrilocal society. Curr. Anthropol. 58, 114–123. Pew Research Center, 2013. At Grandmother's House We Stay. Pew Research Center, Washington, D.C. Pilgrim, N.A., Ahmed, S., Gray, R.H., Sekasanvu, J., Lutalo, T., Nalugoda, F., et al., 2014. Family structure effects on early sexual debut among adolescent girls in Rakai, Uganda. Vulnerable Child. Youth Stud. 9, 193–205. Pilkauskas, N.V., 2012. Three-generation family households: differences by family structure at birth. J. Marriage Fam. 74, 931–943. Pilkauskas, N.V., 2014a. Breastfeeding initiation and duration in coresident grandparent, mother and infant households. Matern. Child Health J. 18, 1955–1963. Pilkauskas, N.V., 2014b. Living with a grandparent and parent in early childhood: associations with school readiness and differences by demographic characteristics. Dev. Psychol. 50, 2587–2599. Pilkauskas, N.V., Cross, C., 2018. Beyond the nuclear family: trends in children living in shared households. Demography 55, 2283–2297. Pilkauskas, N., Dunifon, R., 2016. Understanding grandfamilies: characteristics of grandparents, nonresident parents, and children. J. Marriage Fam. 78, 623–633. Pilkauskas, N., Martinson, M., 2014. Three-generation family households in early childhood: comparisons between the United States, the United Kingdom, and Australia. Demogr. Res. 30, 1639–1652. Pittman, L., 2007. Grandmothers' involvement among young adolescents growing up in poverty. J. Res. Adolesc. 17, 89–116. Pittman, L., Boswell, M., 2007. The role of grandmothers in the lives of preschoolers growing up in urban poverty. Appl. Dev. Sci. 11, 20–42. Poehlmann, J., Park, J., Bouffiou, L., Abrahms, J., Shlafer, R., Hahn, E., 2008. Representations of family relationships in children living with custodial grandparents. Attach. Hum. Dev. 10, 165–188. Pong, S.-L., Chen, V., 2010. Co-resident grandparents and grandchildren's academic performance in Taiwan. J. Comp. Fam. Stud. 41, 111–129. Profe, W., Wild, L., 2017. Mother, father, and grandparent involvement: associations with adolescent mental health and substance use. J. Fam. Issues 38, 776–797. Pulgaron, E., Marchante, A., Agosto, Y., Lebron, C., Delamater, M., 2016. Grandparent involvement and children's health outcomes: the current state of the literature. Fam. Syst. Health 34, 260–269. Radl, J., Salazar, L., Cebolla-Boado, H., 2017. Does living in a fatherless household compromise educational success? A comparative study of cognitive and non-cognitive skills. Eur. J. Popul. 33, 217–242. Ragsdale, G., 2004. Grandmothering in Cambridgeshire, 1770-1861. Hum. Nat. 15, 301–317. Razzaque, A., Hossain, M.A., DaVanzo, J., Hoque, M.E., Alam, N., Bhuiya, A., et al., 2014. Effect of maternal mortality on survival of under-five children. Asian Popul. Stud. 10, 60–74. Renny, P.P., Jayasankara, R., 2016. Influence of grandparents on the emotional intelligence of early adolescents in Kerala. J. Indian Acad. Appl. Psychol. 42, 370–373. Reynolds, S., Fernald, L.C., Deardorff, J., Behrman, J., 2018. Family structure and child development in Chile: a longitudinal analysis of household transitions involving fathers and grandparents. Demogr. Res. 38, 1777–1814. Rhodes, K., Chan, F., Prichard, I., Coveney, J., Ward, P., Wilson, C., 2016. Intergenerational transmission of dietary behaviours: a qualitative study of AngloAustralian, Chinese-Australian and Italian-Australian three-generation families. Appetite 103, 309–317. Rillamas-Sun, E., Harlow, S.D., Randolph Jr., J.F., 2014. Grandmothers' smoking in pregnancy and grandchildren's birth weight: comparisons by grandmother birth cohort. Matern. Child Health J. 18, 1691–1698. Rogers, E., Bell, L., Mehta, K., 2019. Exploring the role of grandparents in the feeding of grandchildren aged 1-5 years. J. Nutr. Educ. Behav. 51, 300–306. Rwenge, M., 2000. Sexual risk behaviors among young people in Bamenda, Cameroon. Int. Fam. Plan. Perspect. 26, 118–123. Ryan, L.M., Lindstrom Johnson, S., Jones, V., Fein, J.A., Cheng, T.L., 2019. Is household composition associated with repeat fight injuries in adolescents living in urban neighborhoods? J. Interpers. Violence 886260519829768. Sata, M., Yamagishi, K., Sairenchi, T., Ikeda, A., Irie, F., Watanabe, H., et al., 2015. Impact of caregiver type for 3-year-old children on subsequent between-meal eating habits and being overweight from childhood to adulthood: a 20-year follow-up of the Ibaraki Children's Cohort (IBACHIL) study. J. Epidemiol. 25, 600–607. Schatz, E., Seeley, J., 2015. Gender, ageing and carework in East and Southern Africa. Glob. Public Health 1–16. Schmeer, K.K., 2013. Family structure and child anemia in Mexico. Soc. Sci. Med. 95, 16–23. Schrijner, S., Smits, J., 2018a. Grandmothers and children’s schooling in Sub-Saharan Africa. Hum. Nat. 29, 65–89. Schrijner, S., Smits, J., 2018b. Grandparents and children’s stunting in Sub-Saharan Africa. Soc. Sci. Med. 205, 90–98. Schwartz, R., Vigo, A., de Oliveira, L.D., Justo Giugliani, E.R., 2015. The effect of a probreastfeeding and healthy complementary feeding intervention targeting adolescent mothers and grandmothers on growth and prevalence of overweight of preschool children. PLoS One 10, e0131884. Sear, R., 2008. Kin and child survival in rural Malawi : are matrilineal kin always beneficial in a matrilineal society? Hum. Nat. 19, 277–293. Sear, R., Steele, F., McGregor, I.A., Mace, R., 2002. The effects of kin on child mortality in rural Gambia. Demography 39, 43–63. Seay, D.M., Jahromi, L.B., Umana-Taylor, A.J., Updegraff, K.A., 2016. Intergenerational transmission of maladaptive parenting strategies in families of adolescent mothers: effects from grandmothers to young children. J. Abnorm. Child Psychol. 44, 1097–1109. Sharer, M., Cluver, L., Shields, J., 2015. Mental health of youth orphaned due to AIDS in South Africa: biological and supportive links to caregivers. Vulnerable Child. Youth Stud. 10, 141–152. Sharma, M., Kanani, S., 2006. Grandmothers' influence on child care. Indian J. Pediatr. 73, 295–298. Shaw, D.J., Warren, T.B., Johnson, M.E., 2019. Family structure and past-30 day opioid misuse among justice-involved children. Subst. Use Misuse 54, 1226–1235. Shea, B.J., Reeves, B.C., Wells, G., Thuku, M., Hamel, C., Moran, J., et al., 2017. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or nonrandomised studies of healthcare interventions, or both. Brit. Med. J. 358, j4008. Sheppard, P., Sear, R., 2016. Do grandparents compete with or support their grandchildren? In Guatemala, paternal grandmothers may compete, and maternal grandmothers may cooperate. R. Soc. Open Sci. 3, 160069. Shi, J., Wang, L., Yao, Y., Su, N., Zhao, X., Chen, F., 2017a. Family impacts on self-esteem in Chinese college freshmen. Front. Psychiatry 8, 279. Shi, J., Wang, L., Yao, Y., Su, N., Zhao, X., Zhan, C., 2017b. Family function and selfesteem among Chinese university students with and without grandparenting 31 Social Science & Medicine 239 (2019) 112476 A.F.A. Sadruddin, et al. experience: moderating effect of social support. Front. Psychol. 8, 886. Smith, K., 2018. Parental substance use in New Hampshire: who cares for children? In: The Casey School of Public Policy at the University of New Hampshire, pp. 1–6. Smith, C., Beitranm, A., 2003. The role of federal policies in supporting grandparents raising grandchildren families: the case of the U.S. J. Intergener. Relat. 1, 2003. Smith, G., Palmieri, P., 2007. Risk of psychological difficulties among children raised by custodial grandparents. Psychiatr. Serv. 58, 1303–13010. Smith, G., Palmieri, P., Hancock, G., Richardson, R., 2008. Custodial grandmothers' psychological distress, dysfunctional parenting, and grandchildren's adjustment. Int. J. Aging Hum. Dev. 67, 327–357. Smith, G., Cichy, K., Montoro-Rodriquez, J., 2015. Impact of coping resources on the wellbeing of custodial grandmothers and grandchildren. Fam. Relat. 64, 378–392. Smith, G., Hayslip, B., Hancock, G., Merchant, W., Montoro-Rodriguez, J., Strieder, F., 2018a. The family stress model as it applies to custodial grandfamilies: a cross validation. J. Child Fam. Stud. 27, 505–521. Smith, G.C., Hayslip, B., Hancock, G.R., Strieder, F.H., Montoro-Rodriguez, J., 2018b. A randomized clinical trial of interventions for improving well-being in custodial grandfamilies. J. Fam. Psychol. 32, 816–827. Snopkowski, K., Sear, R., 2015. Grandparental help in Indonesia is directed preferentially towards needier descendants: a potential confounder when exploring grandparental influences on child health. Soc. Sci. Med. 128, 105–114. Soldateli, B., Vigo, A., Giugliani, E.R., 2016. Effect of pattern and duration of breastfeeding on the consumption of fruits and vegetables among preschool children. PLoS One 11, e0148357. Song, J., Ma, C., Gu, C., Zuo, B., 2018. What matters most to the left-behind children's life satisfaction and school engagement: parent or grandparent? J. Child Fam. Stud. 27, 2481–2490. Sun, Y., Jiang, N., 2017. The effect of grandparents’ co-parenting on young children’s personality and adaptation: Chinese three-generation families. Asian Soc. Sci. 13, 7–15. Sun, Y., Li, Y., 2013. Alternative households, structural changes, and cognitive development of infants and toddlers. J. Fam. Issues 35, 1440–1472. Sun, M., Xue, Z., Zhang, W., Guo, R., Hu, A., Li, Y., et al., 2017. Psychotic-like experiences, trauma and related risk factors among “left-behind” children in China. Schizophr. Res. 181, 43–48. Susiloretni, K.A., Hadi, H., Prabandari, Y.S., Soenarto, Y.S., Wilopo, S.A., 2015. What works to improve duration of exclusive breastfeeding: lessons from the exclusive breastfeeding promotion program in rural Indonesia. Matern. Child Health J. 19, 1515–1525. Susin, L.R., Giugliani, E.R., Kummer, S.C., 2005. Influence of grandmothers on breastfeeding practices. Rev. Saude Publica 39, 141–147. Sutchritpongsa, S., 2016. Unintentional poisoning in children: analysis for significant risk factors. Siriraj Med. J. 68, 17–22. Tang, W., Wang, G., Hu, T., Dai, Q., Xu, J., Yang, Y., et al., 2018. Mental health and psychosocial problems among Chinese left-behind children: a cross-sectional comparative study. J. Affect. Disord. 241, 133–141. Tanskanen, A.O., 2013. The association between grandmaternal investment and early years overweight in the UK. Evol. Psychol. 11, 417–425. Tanskanen, A.O., Danielsbacka, M., 2012. Beneficial effects of grandparental involvement vary by lineage in the UK. Personal. Individ. Differ. 53, 985–988. Tanskanen, A.O., Danielsbacka, M., 2017. Association between grandparental co-residence and early childhood injury in the UK. Child Indic. Res. 10, 825–837. Tanskanen, A.O., Danielsbacka, M., 2018. Multigenerational effects on children's cognitive and socioemotional outcomes: a within-child investigation. Child Dev. 89, 1856–1870. Tenkorang, E.Y., Adjei, J.K., 2015. Household living arrangements and transition to sexual debut among young people in Ghana. Sex. Educ. 15, 1–18. Thornberry, T.P., Krohn, M.D., Freeman-Gallant, A., 2006. Intergenerational roots of early onset substance use. J. Drug Issues 36, 1–28. Titiloye, M., Brieger, W., 2009. Infant weaning methods: an intergenerational view from Igbo-Ora Nigeria. Appl. Res. Eval. 29, 323–334. United States Senate, 2018. Special Committee on AGING: Bill to Support Grandparents Raising Grandchildren as Opioid Epidemic Increases Their Numbers. Urita, Y., Watanabe, T., Kawagoe, N., Takemoto, I., Tanaka, H., Kijima, S., et al., 2013. Role of infected grandmothers in transmission of Helicobacter pylori to children in a Japanese rural town. J. Paediatr. Child Health 49, 394–398. U.S. Census Bureau, 2018. America's Families and Living Arrangements: 2018. Van Heerden, A., Wild, L.G., 2018. Grandparent support and mental and behavioural health in middle childhood. Soc. Dev. 27, 366–380. Wang, P., Song, P., Chen, C.P., Liang, J., Jian, F.F., Zhang, J.B., 2009. [A comparative study on psychological and behavioral development in children aged 1 to 3 years fostered by grandparents and by parents]. Chin. J. Contemp. Pediatr. 11, 1006–1007. Wang, C.D., Hayslip Jr., B., Sun, Q., Zhu, W., 2019. Grandparents as the primary care providers for their grandchildren: a cross-cultural comparison of Chinese and U.S. Samples. Int. J. Aging Hum. Dev 91415018824722. Weiss, B., Dang, M., Trung, L., Nguyen, M.C., Thuy, N.T., Pollack, A., 2014. A nationallyrepresentative epidemiological and risk factor assessment of child mental health in Vietnam. Int. Perspect. Psychol. 3, 139–153. Xie, H., Caldwell, L.L., Loy, S., Robledo, M., 2018. A qualitative study of Latino grandparents’ involvement in and support for grandchildren’s leisure time physical activity. Health Educ. Behav. 45, 781–789. Xing, S., Liang, X., Yue, J., Wang, Z., 2016a. Multiple attachment relationships and the impacts on children’s socio-emotional development under the background of grandmother co-parenting. Acta Psychol. Sin. 48, 518–528. Xing, S., Zhou, Q., Archer, M., Yue, J., Wang, Z., 2016b. Infant temperamental reactivity, maternal and grandparental sensitivity: differential susceptibility for behavior problems in China. Early Hum. Dev. 101, 99–105. Xu, H., 2019. Physical and mental health of Chinese grandparents caring for grandchildren and great-grandparents. Soc. Sci. Med. 229, 106–116. Xu, L., Munir, K., Kanabkaew, C., Le Coeur, S., 2017. Factors influencing antiretroviral treatment suboptimal adherence among perinatally HIV-infected adolescents in Thailand. PLoS One 12, e0172392. Xue-Yan, Z., Dong-Mei, L., Dan-Dan, X., Le-Shan, Z., 2016. Obese Chinese primary-school students and low self-esteem: a cross-sectional study. Iran. J. Pediatr. 26 (4), e3777. Young, K., Duncanson, K., Burrows, T., 2018. Influence of grandparents on the dietary intake of their 2-12-old grandchildren: a systematic review. Nutr. Dietics 0, 1–16. Yue, A., Zhang, N., Liu, X., Tang, L., Luo, R., Yang, M., et al., 2018. Do infant feeding practices differ between grandmothers and mothers in rural China? Evidence from rural shaanxi province. Fam. Community Health 41, 233–243. Yun, E.K., Shin, S.H., 2013. Comparison of the factors influencing young adolescents' aggression according to family structure. J. Korean Acad. Nurs. 43, 321–330. Zhang, Y., Wei, M., Zhang, Y.Q., Shen, N.P., 2013. Impact of family management on psychological and behavioral problems in children with chronic diseases. J. Shanghai Jiaot. Univ. 33, 303–309. Zhang, N., Becares, L., Chandola, T., Callery, P., 2015. Intergenerational differences in beliefs about healthy eating among carers of left-behind children in rural China: a qualitative study. Appetite 95, 484–491. Zhao, J., Zhang, T., Lin, L., 2016. Grandparent-child cohesion and depression in leftbehind children: the mediating role of cognitive appraisals for left-behind hassles. Chin. J. Clin. Psychol. 24, 1092–1097. Ziol-Guest, K., Dunifon, R., 2014. Complex living arrangements and child health: examining family structure linkages with children’s health outcomes. Family Relat. 63, 424–437. Zong, X.N., Li, H., Zhang, Y.Q., 2015. Family-related risk factors of obesity among preschool children: results from a series of national epidemiological surveys in China. BMC Public Health 15, 927. 32