Global S&T Development Trend Analysis Platform of Resources and Environment
| DOI | 10.1126/science.abg0132 |
| The impact of early childhood interventions on mothers | |
| David K. Evans; Pamela Jakiela; Heather A. Knauer | |
| 2021-05-21 | |
| 发表期刊 | Science
 |
| 出版年 | 2021 |
| 英文摘要 | Over 250 million children in low- and middle-income countries (LMICs) are at risk of failing to meet their developmental potential, primarily because of a lack of adequate nutrition and stimulation in early childhood ([ 1 ][1]). Well-designed early childhood development (ECD) interventions can have substantial impacts on children's physical, cognitive, and socioemotional development, as well as their eventual schooling attainment, wages, and other outcomes ([ 1 ][1]–[ 3 ][2]). Although the potential indirect effects of interventions on mothers and other household members are generally acknowledged, few studies explicitly quantify outcomes related to labor market activities, health, or well-being of household members other than young children. This may lead policy-makers to overinvest in programs that impose substantial costs on women and to underinvest in those that improve women's well-being. Systematically ignoring impacts on specific subgroups—particularly vulnerable groups such as women in LMICs—risks exacerbating inequalities in the name of evidence-based policy.
Many ECD interventions have important implications for mothers and other caregivers. Center-based care may give mothers more time for outside labor force opportunities, whereas parent education and home visiting programs impose on mothers' time. Parent education programs may improve maternal mental health by equipping mothers with skills to better handle stressful situations, but they may also change parents' beliefs about the extent to which they are taking adequate care of their children. Child-targeted cash transfer programs may reduce stress by loosening household budget constraints.
Research examining the costs and benefits of different approaches to ECD and childcare rarely highlights the value of women's time or measures the costs and benefits of programs that change women's time use. Even if the primary objective of an ECD intervention is to improve outcomes for children, failing to measure the impact on women's time and other maternal outcomes implicitly sets their value at zero.
We examined 3716 studies from databases that cover medical, psychological, economic, and other social science research, and we supplemented the results of this systematic review with 27 additional papers that met our inclusion criteria. We identified a total of 478 studies published between 2005 and 2019 that evaluated ECD-related interventions in LMICs using an experimental or quasi-experimental research design [see supplementary materials (SM) for details on all analyses].
![Figure][3]
A growing, but still too narrow, body of evidence
The number of quantitative studies measuring the impact of early childhood development (ECD) interventions in low- and middle-income countries (LMICs) (light blue), and the number of those studies that report nonparenting outcomes for mothers (blue), have both grown over time.
GRAPHIC: N. DESAI/ SCIENCE
Of those studies, the overwhelming majority (91%) include either outcomes for children aged 0 to 5 years or parenting outcomes relevant for young children. However, only 22% (105 studies) report any mother-specific outcomes that are not exclusively focused on parenting practices. Though both the number of ECD studies and the proportion reporting impacts on mothers have grown over time, the latter proportion remains quite small (see the first figure). Only 19 studies (4%) examined maternal labor market outcomes, and only six (1%) report outcomes related to other aspects of women's empowerment. Studies documenting impacts on women's health are slightly more common (though still relatively rare): 55 studies (12%) report impacts on maternal mental health, while 47 studies (10%) report impacts on women's physical health.
A further 23 studies (5%) report any household-level outcomes beyond parenting practices (e.g., food security or sanitation practices). Only 12 studies (3%) report any father-specific outcomes—including father-specific measures of parenting practices—and only seven (1%) report impacts on older siblings in middle childhood or adolescence. In total, only 119 studies (25%) report outcomes related to the labor market activities, health, or well-being of household members other than young children.
This pattern of selective reporting might make sense if ECD interventions do not affect older household members: If all relevant costs and benefits of programs are captured in child and parenting outcomes, then there is no reason to report impacts on other household members. However, ECD interventions like childcare have obvious, direct implications for household labor supply, whereas many other interventions rely on parents to mediate impacts by changing their behavior (for example, by attending parent education classes or engaging in more stimulating activities with young children). Thus, there is a strong theoretical basis for measuring and reporting impacts on mothers, in particular—even when these impacts are null or imprecise. Moreover, among those studies that measure effects on mothers and other caregivers, many report statistically significant impacts.
Women—particularly mothers but also grandmothers, older sisters, and other relatives—do most of the childcare in most societies ([ 4 ][4], [ 5 ][5]), and the theoretical and empirical relationship between parenting effort, access to childcare, and female labor force participation is well documented in both low-income and high-income settings ([ 6 ][6]–[ 8 ][7]). Nevertheless, only 4% of the impact evaluations in our sample report impacts on women's labor force participation, income, or time use. Fortunately, there is substantial overlap across studies in terms of both the interventions being evaluated and the outcomes reported, making it possible to draw tentative conclusions from the existing evidence base (see the second figure)—though caution is clearly warranted, particularly in regard to sample selection if null results are less likely to be reported.
All four studies that measure the impact of access to daycare on women's employment (in Brazil, Ecuador, Kenya, and Nicaragua) report some positive impacts—from a 5 percentage point increase in the employment of a child's primary caregiver in Brazil (though in that setting the child's primary caregiver is often a grandmother or adult sister rather than the child's mother) to a 31 percentage point increase in maternal employment in Ecuador. Differences in the size of effects across settings may depend on differences in job opportunities and other contextual factors (see SM for details on all studies analyzed for this article).
Existing evidence also suggests that access to daycare has a positive impact on women's income and overall household income. In Kenya, access to daycare increased mothers' income. In Brazil, access to daycare increased household income and had a marginally significant impact on the income of a child's primary caregiver, though it did not increase maternal income significantly. In Ecuador, childcare centers had a significant positive impact on both mother's income and the income of the household head. All four studies estimating the impact of preschool on women's labor force participation also report some weakly positive impacts, though the evidence is less robust than in the case of access to daycare.
Thus, the existing evidence base suggests that expanding access to daycare and preschool is likely to increase women's employment and household income in many settings—but less than half of the evaluations of such interventions in our sample report impacts on these outcomes. Many evaluations may be ignoring important program impacts. Moreover, whereas interventions such as daycare that reduce women's carework burden tend to increase female labor force participation, ECD programs that encourage parents to increase their parenting effort might have the opposite effect. A home visiting intervention in Ecuador decreased maternal labor force participation, although a similar program in South Africa did not. Though interventions involving home visits are increasingly common, impacts on mothers' time use and labor supply are almost never measured—making it impossible to know whether these programs alter women's domestic burden. Studies that do not measure the impacts of ECD interventions on women's time use and labor supply cannot fully account for the costs and benefits of such policy interventions.
![Figure][3]
Targeting children, affecting mothers: labor force participation
Estimated impacts (and 95% confidence intervals) on mother or other caregiver labor force participation are shown from studies of different interventions in low- and middle-income countries. Although too few studies measure these outcomes, those that do suggest that daycare and preschool interventions tend to have either no impact or a positive impact on women's labor force participation. For details on sources and estimates, see the supplementary materials.
GRAPHIC: N. DESAI/ SCIENCE
ECD interventions have the potential to affect maternal mental health in several ways. Because women do most of the childcare in LMIC contexts, any intervention targeting young children has the potential to bring mothers into contact with the implementing partners (e.g., child development specialists or government health care workers) and other mothers participating in the intervention. These interactions may reduce feelings of isolation and strengthen social support networks. This is particularly true of home visiting interventions and parent education classes, which work by engaging primary caregivers in regular interactions with individuals outside their own household. At the same time, these interventions impose on mothers' time, which could increase stress or reduce time available for income generation.
Interventions like cash transfers are often targeted to women, sometimes with the explicit objective of increasing mothers' autonomy and decision-making power, whereas daycare and other interventions that influence women's labor force participation may improve women's bargaining power and strengthen their social networks outside the home.
Existing empirical evidence supports the hypothesis that many ECD interventions in LMICs have positive impacts on mothers' mental health and well-being. In our sample of 478 impact evaluations of ECD interventions, 55 (12%) report impacts on any outcomes related to maternal mental health—and 27 of those report positive (in the sense of improving women's well-being), statistically significant impacts. Only three find negative, statistically significant impacts.
Group-based parent education interventions reduced maternal depression in Bangladesh, Guatemala, Pakistan, South Africa, Uganda, and Zambia; and home visits from child development professionals reduced maternal depression in Bangladesh, Jamaica, Iran, and Pakistan. In other settings, similar interventions improved maternal self-confidence and parental efficacy—even when impacts on overall mental health were not statistically significant. Evaluations of these interventions were also more likely to report impacts on maternal mental health: 80% of studies reporting impacts on maternal mental health are evaluations of either home visiting programs or group-based parent education, even though these types of interventions do not account for 80% of all studies in our sample. Studies of cash transfers rarely report impacts on maternal mental health, and those that do have typically not found statistically significant positive impacts—though cash transfers have been shown to improve overall well-being in other contexts ([ 9 ][8])—and evaluations of other ECD interventions (including daycare and preschool programs) that report impacts on maternal mental health are exceedingly rare.
Group-based parent education programs and interventions involving home visits are both common, and it is impossible to tell whether the papers that report maternal mental health outcomes are representative of the overall distribution of impacts. Caution is warranted—it should not be concluded that these programs automatically improve women's mental health. However, the number of impact evaluations of home visits and group-based parenting classes that find positive impacts on maternal mental health is higher than would be expected if these results were spurious. Evaluations that do not capture potential impacts on maternal mental health cannot fully evaluate program costs and benefits.
Although female labor force participation and mental health are both entwined with women's empowerment, only six studies evaluate the impacts of ECD interventions on women's autonomy or decision-making power within the household—but five of the six find some evidence of positive impacts. In contrast to the cases discussed above, there is no evidence that studies of one specific class of intervention are more likely to report impacts on empowerment: The six studies that measure empowerment include evaluations of a program to treat postpartum depression, an intervention intended to promote fathers' involvement in child-rearing, a cash transfer program targeting parents of young children, and a program providing vouchers for daycare.
This range of interventions—all of which appear to increase women's financial autonomy or involvement in household decision-making—highlights the connections between women's childcare responsibilities, their sense of social support, their financial independence and agency, and their mental well-being. Evaluations that ignore these impacts cannot trace out the mechanisms that explain why interventions work in some contexts and not in others.
Households—and especially mothers—invest substantial time and energy in their young children, and ECD interventions often have large impacts on child-rearing1 practices. Research that ignores impacts on caregivers can only provide a partial characterization of program effects. Relatively few evaluations of ECD interventions in LMICs report impacts on mothers, and even fewer report impacts for other caregivers such as fathers or older siblings. Beyond interventions intended to boost child outcomes, other studies demonstrate the adverse impacts of negative experiences like early childhood illnesses on parents and older siblings ([ 10 ][9]).
The expected link between ECD interventions and women's outcomes varies across interventions. Providing daycare or preschool has obvious implications for women's labor force participation. Purely medical interventions for young children—which we have excluded from our review—have less obvious implications, as do some health and nutrition interventions such as vitamin supplements. However, many studies fall between those two extremes. Specifically, interventions such as parenting education intended to change mothers' behavior are increasingly common. When an intervention's theory of change involves changing the behavioral practices of mothers (or other caregivers), it makes sense to ask how mothers will be affected by treatment. Yet, many researchers who study ECD interventions are primarily interested in improving outcomes for children, and this may predispose them to focus on child rather than mother outcomes.
There can be no evidence-based policy regarding caregiving interventions without evidence on the caregivers. Our findings point to the urgent need to routinely incorporate measures of caregivers' productivity and well-being into evaluations of ECD interventions. They also suggest the importance of evaluating ECD programs against multiple objectives. A program that delivers gains for both mothers and children may be more cost-effective than a program that delivers similar gains for children alone.
This has real-world policy implications. A 2015 study published by the Inter-American Development Bank found much higher benefit-cost ratios for home visiting programs relative to daycare programs, but only factored benefits to children into the calculations ([ 11 ][10]). Incorporating benefits to women shifts those calculations, even if it does not overturn them ([ 12 ][11], [ 13 ][12]). These kinds of calculations can influence national investment programs.
Ultimately, the evidence from those few studies that do measure impacts of ECD interventions on women suggests that many programs that are good for children also yield benefits for women's financial well-being, mental health, and overall empowerment. The case for ECD interventions may be even stronger than previously believed and may open the door to financing these programs from a wider range of sources. More of this measurement may shift emphasis from ECD interventions that are best for children to those that are best for children and for their caregivers. Policy-makers can only identify these trade-offs and interdependencies if they have information about all the gains that programs deliver.
[science.sciencemag.org/content/372/6544/794/suppl/DC1][13]
1. [↵][14]1. M. M. Black et al
., Lancet 389, 77 (2017).
[OpenUrl][15][CrossRef][16][PubMed][17]
2. 1. P. L. Engle et al
., Lancet 378, 1339 (2011).
[OpenUrl][18][CrossRef][19][PubMed][20][Web of Science][21]
3. [↵][22]1. P. Gertler et al
., Science 344, 998 (2014).
[OpenUrl][23][Abstract/FREE Full Text][24]
4. [↵][25]1. S. B. Hrdy
, Mothers and Others: The Evolutionary Origins of Mutual Understanding (Belknap, 2009).
5. [↵][26]1. D. F. Lancy
, The Anthropology of Childhood: Cherubs, Chattel, Changelings (Cambridge Univ. Press, Cambridge, ed. 2, 2014); [www.cambridge.org/core/books/anthropology-of-childhood/B34D307F81527FC3C91AE9D0B02D48D7][27].
6. [↵][28]1. J. J. Heckman
, “Effects of Child-Care Programs on Women's Work Effort” (c3687, National Bureau of Economic Research, 1974); [www.nber.org/books-and-chapters/marriage-family-human-capital-and-fertility/effects-child-care-programs-womens-work-effort][29].
7. 1. E. U. Cascio
, J. Hum. Resour. 44, 140 (2009).
[OpenUrl][30]
8. [↵][31]1. D. Del Boca
, “Child care arrangements and labor supply” (2015); |
| 领域 | 气候变化 ; 资源环境 |
| URL | 查看原文 |
| 引用统计 | |
| 文献类型 | 期刊论文 |
| 条目标识符 | http://119.78.100.173/C666/handle/2XK7JSWQ/328811 |
| 专题 | 气候变化 资源环境科学 |
| 推荐引用方式 GB/T 7714 | David K. Evans,Pamela Jakiela,Heather A. Knauer. The impact of early childhood interventions on mothers[J]. Science,2021. |
| APA | David K. Evans,Pamela Jakiela,&Heather A. Knauer.(2021).The impact of early childhood interventions on mothers.Science. |
| MLA | David K. Evans,et al."The impact of early childhood interventions on mothers".Science (2021). |
| 条目包含的文件 | 条目无相关文件。 | |||||
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论