Microsoft word - 4th paper final english brito.doc

CHILD HEALTH AND EDUCATION, 1 (2), 112-133 Simon Fraser University, British Columbia, Canada ISSN 1911-7758 Early Intervention and Caregiving: Evidence from the Uganda Nutrition and Early Child Development Program Pia Rebello Britto, Ph.D. Child Study Center, Yale University Patrice Engle, Ph.D. California Polytechnic Institute Harold Alderman, Ph.D. It was hypothesized that the World Bank assisted project on Nutrition and Early Child Development in Uganda would have an impact on parenting behaviors and attitudes about child development. This large-scale World Bank-assisted program operated in 35 districts during the period from 2000 to 2003. An experimental pre-post comparison group design (N=2250 families) was used to compare behaviors and attitudes of those who lived in the intervention area from those who did not. Results showed that at baseline there were few differences in attitudes and practices between groups. However at the final assessment, intervention mothers reported significantly more behaviors and positive attitudes supporting children's learning and development compared to the control group. The intervention group was also more likely to report attitudes supporting father involvement. The data suggest that parenting behaviors and attitudes regarding stimulation and support for children's development and learning can be changed through a program intervention at scale We would like to acknowledge the funding and support for data analysis provided by the World Bank. Correspondence concerning this article should be addressed to Pia Rebello Britto, Child Study Center, Yale University, 230 South Frontage Road, New Haven, CT 06520, Tel (203-785-3903); fax (203-785-7402) e-mail [email protected] Early intervention and caregiving 113 During the early years children make great strides across all domains of development (McCartney & Phillips, 2006; Shonkoff & Phillips, 2000). The near universal acceptance of the importance of caregiving cannot be disputed as parents and caregivers are the primary providers of stimulating and supportive experiences associated with these developments (Bradley & Corwyn, 2005; Britto, Fuligni, & Brooks-Gunn, 2002; Richter, 2004; Rogoff, 2003; Werner & Smith, 2001; Whiting & Edwards, 1988). Early childhood intervention programming has been fueled by the converging streams of knowledge that early childhood is a period of rapid growth and development, that both biological and social processes in the early years are critical, and that caregiving has a significant effect on children's growth and development beyond the effects of socio-economic status (Bradley et al., 2005). However much of this evidence has been garnered from developed countries (Barnett, 1995; Shonkoff & Phillips, 2000), with developing country data just emerging (Engle, Black, Behrman, et al., 2007). Engle et al (in press) in this journal outlined the significant associations between early learning and later development, and the size and extent of the loss of developmental potential in the developing world. The paper notes that even though nutritional deficiencies are being recognized, much less is known about strategies to improve children's learning opportunities in the earliest years in resource-poor countries and early child development interventions are rarely evaluated. Existing data suggests that early childhood development (ECD) programs that provide comprehensive services, e.g., some combination of health care, nutrition, and early stimulation have pervasive and sustained beneficial impacts on children's well-being, compared to single-service programs (Alderman, Behrman, & Hoddinott, 2004; Young, 1997). Albeit, how these programs should be combined and the best strategies for promoting early stimulation at scale are still unclear. Engle et al. (2007) identified what appeared to be characteristics of successful programming from 20 evaluations of early child development programs in developing countries. These included sufficient duration, provision of direct services to children, intensity, quality, comprehensiveness, targeting to disadvantaged children, and child's age at beginning the project, but the evidence for the importance of each of these is still limited in developing countries. The evaluation of an integrated nutrition and early child development program supported by a World Bank loan to the Government of Uganda for five years, from 1998 to 2003, provides an opportunity to examine the effectiveness of a comprehensive large scale intervention on selected outcomes related to early child development. The Uganda Nutrition and Early Child Development (NECD) program, called the CHILD program, possibly the largest ECD initiative in East Africa. The intervention program was designed to enhance the capacity of families and communities to foster the learning and development of children under 6 years of age (preschool or at age of school entry) through enhancing growth and development. Specifically, communities were helped to organize services for children under 6 years through growth monitoring and promotion activities, strengthening the capacity of families and communities through sensitization, education and skills training on early childhood development, nutrition and health, enhancing access to child health services through six-monthly Child Health Days, and to providing support to communities through community grants for nutrition, savings groups, income generation, and early child development centers. The multi-site longitudinal evaluation of the NECD project was based on randomized experimental 114 Britto, Engle, & Alderman design. The present study is an investigation of one proposed outcome of the intervention -- family caregiving behaviors and attitudes. Uganda is a challenging environment vis-à-vis for support early child development. While Uganda is approaching universal primary school initiation; there are questions of quality of schooling in many areas; additionally, there are many parts of the country that cannot ensure adequate health and nutrition of their children. For example, at the time of the study's initiation, over one third of children were stunted, a rate above the global norm (UNICEF, 2006). Stunting is of particular concern because stunting has been shown to significantly affect cognitive development (Walker et al., 2007). Despite poverty and high rates of malnutrition, most Ugandan families' value education; they want their children to attend school. To illustrate this interest, when, Uganda initiated a program of free primary education, the rates of attendance increased dramatically. Because the focus of the project was to help families support children's early child development, formative research was conducted prior to the intervention design to assess family attitudes and practices toward "positive parental interaction" to support growth, learning and development, and to determine how best to phrase and communicate the messages of the intervention (Baume, Neema, Kibombo, & Cabanero-Verzosa, n.d.). Some results of the intervention have been reported elsewhere. For instance, Alderman (2007) and Alderman and Engle (in press) report that there were improvements in growth but only for the youngest age group. They also report an increase in exclusive breastfeeding, as well as greater dietary diversity (Britto et al., 2005). Use of health services and coverage with deworming treatment also improved (Alderman, Britto, Engle, & Siddiqi, 2004). A few impacts of the intervention on children's cognitive ability, based on the a Ugandan version of the British Ability Scales, for children 3.5 to 6 years of age have also been reported in an unpublished report (Alderman et al., 2004), and further work is underway. The focus of the present paper is the effects of the intervention on parenting behaviours and attitudes that support young children's learning and overall development, behaviors that are important for and support the learning and development of young children (Bradley & Caldwell, 1995). Evidence that caregiver support and cognitive stimulation and provision of opportunities, materials and experiences will promote and further positive child development is strong, especially during the early childhood years (Bornstein, 2002; Richter, 2004). For example, stimulation of learning has been associated with better cognitive and behavior development in early childhood (Fuligni, Han, & Brooks-Gunn, 2004; Leventhal, Martin, & Brooks-Gunn, 2004). Positive parent children interactions and opportunities for development and learning are considered aspects of stimulating and supportive caregiving, e.g., parent and child book reading, playing, singing together (Britto, Fuligni & Brooks-Gunn, 2002; Custodero, Britto, & Brooks-Gunn, 2003). These effects continue to be seen even after controlling for socio-economic status (Klebanov, Brooks-Gunn & Duncan, 1996). For this intervention, based on the formative research and the developmental literature dimensions of caregiving included support and stimulation for children's development (Bradley, 1994). Specifically 3 aspects of stimulation and parental support are assessed in the present study: daily routines; behaviors that support children's learning and preparedness for school and children's play and every day experiences. In Early intervention and caregiving 115 addition, father attitudes and involvement and with young children; and attitudes about the importance of these caregiving activities for young children's learning and development are also assessed. The role of fathers in providing stimulating caregiving is important for young children's development (Cabrera, Tamis-LeMonda, Bradley et al., 2000). Father involvement in child rearing has been considered from two perspectives, traditional and contemporary. Traditional approaches to understanding father involvement typically include the extent to which the father is the primary source of income for the child's education and well-being, i.e., indirectly involved in the development of the child. Contemporary approaches to father involvement take into consideration the father's direct role in caregiving, e.g., bathing, feeding, interactions, etc. fairly frequently (e.g., on a daily basis; Pruett, 2000). The second approach to father involvement is explored in the present study. Attitudes play a role in influencing caregiving behaviors through the caregivers' beliefs in self-efficacy (their belief that they can be effective in caregiving) and in their role construction (what they perceive as their appropriate role; Bandura, 1997; Hoover-Demsy & Sandler, 1997). Parental attitudes towards their role and belief in their ability to help their children's learning have been linked with parenting behaviors and interactions with their children (Ardelt, & Eccles, 2001; Bandura, Barbaranelli, Caprara, & Pastorelli, 1996). These attitudinal components are particularly important for parental involvement in their children's learning and schooling (Alexander, Entwisle, & Bedinger, 1994; Dauber, Alexander, & Entwisle, 1996). Therefore, finding a change in parental attitudes could have implications for behavioral change in parent's involvement in their children's learning and development. The present study examines effects of a comprehensive early childhood intervention on caregiving behaviors and attitudes related to children's development. We hypothesize that caregivers in the intervention group are more likely to report behaviors related to cognitive stimulation, attitudes in favor of cognitive stimulation, and an increase in the role of fathers with young children at the final assessment. These three areas were specifically targeted in the intervention. It is unlikely that there will be an impact on daily experiences and play activities, since it was not a focus of the intervention and is not generally regarded as a task specifically linked with education and learning. Intervention Description The Uganda Nutrition and Early Child Development Project (NECD) was designed to enhance the capacity of families and communities to foster the growth, learning and development of children under 6 years of age (preschool or at age of school entry). The project had three aims: 1) help communities organize services for children under six years, though growth monitoring and promotion, and establishment of early childcare and education facilities; 2) strengthen the capacity of families and communities through sensitization, education, and skills training on early child development, nutrition, and training for savings and income generation; and 3) provide support to communities 116 Britto, Engle, & Alderman through community grants and incentives. These aims were achieved with three inter-related components: community based interventions and grants, sub-district activities, and national-level initiatives including capacity building, not described here. The most disadvantaged districts were selected based on level of malnutrition, infant mortality, and rate of primary school enrollment. The community based intervention was in two-thirds of all communities in these districts. Under NECD, services were delivered through a decentralized system coordinated by local Non-Governmental Organizations (NGOs) and Community Based Organizations (CBOs). These organizations provided training for child caregivers on relevant caregiving activities, training for resource mobilization, child growth promotion and nutrition counseling. Child growth promotion was provided by two volunteers per community who were provided a bicycle as well as training. In addition, communities under the project had access to a grant set in the course of the project at $1,500 ($1,250 from the Bank loan and the remainder from the district). The community determined the specific intervention to be funded from this grant through a process of participatory community planning. These community level workshops also provided an opportunity for communities to learn more about factors related to early child development. At the parish (sub-district) level "Child Health Days" were held every six months and provided an opportunity to communities to access integrated health services (Alderman et al. 2006). In addition, the child days provided an opportunity for imparting information about health, nutrition, and food security though a range of interactive methods. Finally, an educational campaign was initiated that included messages and information on growth and child development. Key messages in the communication campaign, which used radio, posters, and other local media, recommended improved health seeking behaviour, breastfeeding, complementary feeding, and several messages about early child development: the importance of children's learning through play activities and the importance of fathers in the development of their children. These messages were developed through formative research prior to the study design to assess family attitudes and practices to support growth, learning and development, and to determine how best to phrase the messages (Baume, Neema, Kibombo, & Cabanero-Verzosa, n.d.). The formative study videotaped children in 3 communities, conducted interviews with n =100 men and n =100 women, and held focus group discussions in 17 communities across Uganda. Results suggested that parents have very little time for interacting with young children, and children's intelligence is viewed as largely intrinsic. The role of the parent in preparing children for school focused on obedience and training in respect, and good feeding and care. However, the families also reported that they recognized the importance of positive parenting, of stimulating a child's talking, answering his questions, and giving the child attention, although "attention" referred primarily to physical care. A majority of both mothers and fathers reported that they played with the child most days. Almost half reported telling stories, 66% sang, and 33% said that they read to the child on most days. Parents also reported teaching children something, although it was usually physical care, household chores, or how to behave. The authors note that none of these activities were observed in the video sequences, but perhaps they occurred in the evening, whereas videos were taken during Early intervention and caregiving 117 the day. Although parental teaching may be valued, the formative research did not identify when or how often it is done (Baume et al., nd). A special focus of the formative research was to find ways to expand the father's involvement with young children. This involvement was quite limited, and the recommendation for changing this involvement was to work through community norms and behavior rather to rely on messages to individual families. In response, the intervention included community level activities, a publication in simple language on fathers, and other activities. Therefore, the project was designed to help families understand their role in early childhood development and in particular in their role in improving learning opportunities for young children in the home (Baume et al., nd). Implementation faced several challenges. Although the Project was originally planned to be in operation for five years, many of the activities did not begin until late 2001, resulting in less than two years of exposure before the final assessment. Community grants were not allocated before 2002, and in most cases even later, and some communities did not receive grants at all. There was less funding per district than initially planned. Initially the program was limited to 25 districts due to financial constraints, but the number of districts was increased to 34 in order to obtain parliamentary approval, although the funding was not increased. Evaluation design The multi-site longitudinal evaluation of the NECD program was based on randomized experimental design consisting of three groups (Group A: parenting training and ECD services and experimental delivery of Albendazole for deworming and nutritional services; Group B: parenting training and ECD services only; and Group C: control group with no services; Alderman, et. al., 2003). A baseline assessment was done on a random sample of N=2250 households in 2000 and the same households, and similar aged children were assessed in 2003. Thus the analysis of effects allows a comparison of baseline and outcome measures. The random sample was selected based on a 3 strata (parish, community, and household) cluster sampling technique with random selection at each level (See Table 1). Fifty parishes were selected from a random draw of parishes in the project site and randomly assigned to either Group A which received all ECD services as well as the experimental delivery of Albendazole for deworming at child health days. Group B received all the core ECD services. In addition, for each sub county in the project a control group was selected from a sub county which was not in the project but adjacent to it. All parishes in these sub counties were then listed and a random draw of 25 parishes from the total list selected. Two villages from each parish selected were then chosen, again using a list of all the villages in the parish. This step reduced the number of villages where a household listing would be required. A census of households was then conducted in each sample village and a random draw of households containing at least one child under 6 was selected. A baseline questionnaire was administered to 2250 households (750 in each strata) by the Institute of Public Health (IPH) between January and March 2000. Each household with a child under six was revisited in the same season in 2003. Of the initial 2250 households, a total of 2037 were located for re-interview. Households in the parish 118 Britto, Engle, & Alderman who fostered children who were orphaned or otherwise changed households since the baseline were also included in the 2003 sample. If an original household no longer had a child under 6 years and, thus, not able to contribute information relevant to the early child development evaluation, it was dropped from the survey after the roster was filled in and a replacement household with a suitable aged child was selected (Alderman, et. al., 2004). The sampling approach for the household survey was influenced by the costs and logistics of revising households and the need to design survey instrument, including tests of cognitive development, in the local languages. For these reasons this aspect of the analysis was concentrated on a single region with the understanding that other aspects of the evaluation would be conducted more broadly. As one aspect of the study was to test the administration of Albendazole in Child Days (see Alderman et al. 2006) the selection of the Eastern region for this study was based on the results of a study by the Vector Control Unit of the Ministry of Health which indicated the highest rates of worm loads (number of worms per stool) in the country (cf. Kabatereine et al. 2001). Table 1: Evaluation Study Design All ECD services ECD Services only Community2: 2 per parish Household3: 15 per 1 A Parish in Ugandan local government hierarchy is below the district-level. 2 A community is the equivalent of a village 3 Given the conditions associated with the Civil war and the AIDS epidemic have lead to the definition of a household extending beyond the immediate biological family. Second, from a cultural perspective the extended family plays a key role in socialization practices. In the NECD a household is defined as a unit of people living together under one roof and generally sharing meals. Early intervention and caregiving 119 Study Sample Since the major intervention to effect parental childrearing was the parent education and community improvement components, the two intervention groups were combined for this study. Additionally although deworming, when administered twice a year over a period of several years significantly improved weight gain (Alderman et al., 2006), it had no impact on the British Ability Scales score (Alderman, et. al., 2004). The equivalence of the baseline data were evaluated through comparisons on outcome indicators such as nutritional status, socio-economic status and income levels. No differences were found between the treatment groups and the controls (Alderman et al., 2004; IPH, 2002). Table 2 describes sample characteristics for the control and intervention groups at posttest. On average the mothers had completed 3 years of schooling and on average the father had completed 2 more years, i.e., 5 years of schooling. The two groups were comparable at final assessment for child age, gender, maternal education and paal education. However, there were significant differences in number of possessions, a measure of wealth, at the final assessment that had not been present in the initial assessment. Some changes in socio-economic status from 2000 to 2003 were limited to the project area, but others were seen across all project areas (IPH, 2003). Households in general increased the quality of their housing, access to safe water sources, and labor force participation. In the project areas, there was a small increase in families owning bicycles, and in families raising chickens, both project goals. The baseline survey found that over 36% of children under the age of 5 were moderately and severely malnourished. Food shortages were reported to be common, in the baseline period, 30% of families reported having only one major meal a day for at least 10 weeks. Houses were generally of basic quality, with the majority having grass roofs and dirt floors. Not surprisingly, malnutrition was high; the overall rate of stunting at 39%, fairly consistent across the three groups, is higher than found in many parts of sub-Saharan Africa. HIV/AIDS is an overwhelming issue in Uganda, and one concern is that many of the children may have been without parents because of the effects of the epidemic of HIV/AIDS in the country. Families were asked about the status of the mother. A total of 113 mothers (6.5%) were said to be "away", and 14, or 0.8% were reported to have died. The low number is not surprising in that orphanhood increases with child age, with fewer young children likely to be affected. 120 Britto, Engle, & Alderman Table 2: Selected Sample Characteristics at Baseline and Final Assessment. M(SD)/%(n) M(SD)/%(n) Number of possessions at 1.91 (.98) (n=1493) 1.86 (.91) (n=751) baseline (t=1.31, ns) Number of possessions at 2.87(1.41) (n=1142) 2.60(1.31) (n=635)** final (t=3.86, p<.001) Child Age (months) final 36.2 (16) (n=1243) 35.6 (16) (n=660) Child Gender (Female) final 49 % (n=1243) 51% (n=660) Maternal Education (#yrs. 3.15 (3.20) (n=926) 3.11 (3.04) (n=491) of school) final Father Education (# yrs. of 4.82 (4.05) (n=926) 4.76 (4.16) (n=491) school) final *p<.05; ** p<.01; *** p< .001 Measures Outcome variables in the evaluation included nutritional status, attitudes and beliefs about nutrition, vaccination rates, use of health services, test score on an adaptation of the British Ability Scales (a measure of language and cognitive development, for children aged 3.5 to 6 years), and parental attitudes and practices about early child development. There are nine modules to the household questionnaire addressing various aspects of the home life ranging from housing assets, utilities, labor force participation, etc. The current study focuses on the caregiving attitudes and behaviours. Caregiving attitudes and behaviors were measured through questions to the mother or female head of household in the baseline and final household survey. Questions were asked about a specific child, the child closest to 3 years of age. On average, children were 36 months. Three areas of caregiving were assessed: 1) behaviors that support learning and development; daily routines and caregiving; and daily experiences including play; 2) father involvement (behavior and attitudes) in caregiving; and 3) attitudes toward parents' role in a child's learning and development. The assessment of behaviors that support learning and development was based on aspects of caregiving adapted, in part from the Home Observation for Measurement of the Environment or HOME inventory (Bradley, 1994). HOME inventory is the most widely used assessment of children's home environments in the field of developmental psychology. The HOME is intended to measure the quality and quantity of stimulation and support available to a child in the home environment. Aspects of stimulation and support included in the present study were: Physical caregiving – 1 item (e.g., On the last day you were with the child, did the child receive care of physical needs from you (e.g. bathing, feeding)); Early intervention and caregiving 121 Child involvement in routine household activities – 2 items (Spent time with you doing work of some sort (e.g., sweeping, washing, agriculture; child did errands) Experiential and play activities – 11 items (e.g., spent time with mother/respondent playing, talking & laughing; went outside village; child had purchased or made toys); and Learning activities linked with school preparedness – 3 items (Spent time in learning activities (counting, naming objects, drawing, etc.)). Father involvement was assessed with four questions about his activities with the child on the previous day (e.g., spent time with father playing, laughing and talking), and two questions on attitudes (e.g., Fathers' responsibilities for children are ONLY to provide food and medicine). The final set of 7 questions addressed caregiving attitudes towards their role in preparing children for school (e.g., making toys and playthings for children under 6 helps them to get ready for school; children become more intelligent if they eat a good diet when they are young (under 3). Data Analytical Approach The survey instrument was essentially unchanged from baseline to final assessment. Although the same families were followed from the baseline to the final survey, in the case of the caregiving questions, it was not possible to ask the caregiver questions on the same child, as the developmental tasks would have changed. Therefore in both assessments, the target child was the one closest to the age of 36 months. Although it is possible that the same child would have been selected for both rounds, given the fertility rates in Uganda and the three-year time difference, it is more likely that a younger child was selected for the second round. However, this may not be a major problem, since the focus of the study is on caregiver behavior not child outcomes. Attrition between baseline and follow-up was very low, as refusals were negligible, few households had relocated, moreover Global Position Systems facilitated locating the families that had moved (Alderman, et. al., 2004). Therefore, the analytic strategy involves a cross-sectional comparison of the baseline data in comparison to the comparison of groups in the final study rather than a longitudinal design. Program effects were examined using descriptive and bivariate analyses. Given the categorical type of data, chi-square analyses were conducted. This analytical procedure is most suited for binomial data and observations involving two categories, often times mutually exclusive (Tabchnick & Fidell, 1996). In the present study the categorical predictor variables are intervention and control status. The dependent variables of parental behaviors and attitudes were also coded on a binary scale. The Pearson 2-sided χ2 tests were conducted to determine significance of the results. A primary aim of the NECD program was to build the capacity of families and communities to improve young children's well-being. This section presents the results of 122 Britto, Engle, & Alderman the intervention impact on caregiving behaviors and attitudes analyzed in the current study. Stimulation and Supportive Behaviors for Learning and Development The first set of analyses examined differences between intervention and control families on stimulation and support for learning and development across physical care, daily routines, play and experiential activities, and learning and school preparedness. Table 3 presents results of the maternal behavioral report of behaviors at baseline and at the final assessment, and differences between intervention and control. At baseline a significantly greater percentage of control mothers reported giving physical care for their children (χ2(1) = 4.76, p=.029) whereas at the final assessment, a greater percentage of intervention mothers reported doing daily physical care behaviors (χ2(1) = 5.26, p=.022). In terms of daily routine activities, at baseline a greater percentage of control mothers reported involving their children in their daily routines of sweeping, washing and agriculture (χ2(1) = 9.06, p=.003)) which was reversed at final assessment with a significantly higher percentage of intervention mothers involving their children in daily routines (χ2(1) = 5.02, p=.025). The behaviors which differed most between the baseline and the final assessment were the three related to learning preparedness (spent time in an early childhood education center, spent time in learning activities (counting, naming objects, drawing), and spent time in dancing, drawing, and singing). At baseline, none of these three behaviors differed significantly between groups. However, at the final assessment, there were significant differences in all 3 behaviors, all favoring the intervention group (.χ2(1) = 7.21, p=.005; χ2(1) = 7.92, p=.005; χ2(1) = 5.90, p=.015, respectively) Among the play and experiential activities, only 2 of out of 11 items differed at pretest (more intervention families reported a special place to put children's things, (χ2(1) = 4.025, p=.039), and more intervention families had homemade toys for their child (χ2(1) = 8.02, p=.005)). Two differed at posttest (more intervention families reported taking the child outside village (χ2(1) = 3.81, p=.05), and the home has a special place for this child to play (χ2(1) = 13.91, p=.0001), Overall the number of families who reported homemade toys increased from 14% to 33%, although there was no increase in the number of purchased toys (10% ). Early intervention and caregiving 123 Table 3: Percent caregivers reporting Stimulation and Supportive Caregiving Behaviors at Baseline and Final Assessment Baseline Assessment Final Assessment 2003 Question: In the last day you Intervention Control Intervention Control were with the child, did the child (n=1493 ) (n= 751) (n=1310) do any of these?
Physical care
Did the child receive care of physical needs from you (e.g.
bathing, feeding)
Household activities
Did an errand at someone's request Spent time with you doing work of some sort (e.g., sweeping,
washing, agriculture)?
Learning activities for school
preparedness
Spent time in an early childhood
education centre (2000: asked parents' plans to send child to preschool) Spent time in learning activities (counting, naming objects, drawing, etc.) Spent time dancing and drawing Play and experiential Activities
Spent time listening to songs or stories told to him/her Went outside compound Went outside village Home has a special place to put his/her things Played with other children Played with toys Spent time with you playing, talking & laughing? Played by imitating or copying some of your actions like washing dishes Home has a special place for this child to play Child has toys that were purchased for him/her 124 Britto, Engle, & Alderman Baseline Assessment Final Assessment 2003 Question: In the last day you Intervention Control Intervention Control were with the child, did the child (n=1493 ) (n= 751) (n=1310) do any of these? Child has toys that a parent or other adult made for him/her Note: * <p.05; **<p.01; **<p.001; ***<p.0001 Father Involvement For the category of father involvement (Table 4), differences were limited. The only difference at pretest was that intervention mothers were more likely to agree that fathers need only be responsible for provision (χ2(1) = 19.26, p=.0001), which was not seen at final assessment; and at posttest, mothers were more likely to agree that there are ways that fathers can help their children learn to talk at a young age (χ2(1) =7.02, p=.008) Table 4: Father Involvement: Behavior and Attitudes at Baseline and Final Assessment Baseline Assessment Final Assessment 2003 Question: In the last day you were with the child, did the child do any of these? (n=1493 ) Receive care of physical needs from his/her father (e.g., bathing, feeding) Spent time with father playing, laughing and talking Spent time with his/her father doing work of some sort (e.g., sweeping, washing, agriculture)? There are ways that fathers can help their Not given Not given children learn to talk at a young age (agree) Fathers' responsibilities for children are ONLY to provide food and medicine Fathers don't need to have conversations with children less than 6 yrs old (disagree) Note: * <p.05; **<p.01; **<p.001; ***<p.0001 Caregiving Attitudes Early intervention and caregiving 125 Seven attitude questions assessed the respondent's attitude toward a child's development and preparing her child for school, both cognitively and through dietary support. At baseline, on 2 of these questions the control group was more in agreement with the importance of learning activities than intervention mothers. However, at the final assessment, intervention mothers were clearly more in agreement with the importance of learning activities than control mothers, differing on 5 of the 7 items. Table 5: Caregiving Attitudes supporting Learning at Baseline and Final Assessment (N=2300 at baseline; and N=2100 at Final; % reported) Playing helps a child to learn to think well 94 Playing has nothing to do with how much 71 young children learn (disagree) Parents don't need to find or make anything special for young children to play with (disagree) Making toys and playthings for children under 6 helps them to get ready for school Children become more intelligent if they eat a good diet when they are young (under 3) What you feed a child before the age of two has no effect on how well they do in school (disagree) Teaching children colors and numbers before they go to school is a waste of time (disagree) Note: * <p.05; **<p.01; **<p.001; ***<p.0001 Finally, we report on exposure to the project, to understand the results obtained. The project used a number of communication approaches such as radio, so it was likely that both project and non-project households may have been exposed to the NECD (called CHILD) project. When sample parents were asked if they had heard of the project, 93% in the project area reported having heard of it compared to 46% in the non-project area (χ2 (1) = 538.1, p<.0001). Of those who had heard of the project, 83% of parents in the project area reported face to face contact with the project personal compared to 35% of non- project parents. Not surprisingly, non-project parents who had heard of the program were much more likely to know it through mass media only (65%) than were project parents (17%) (χ2 (1) = 254.8, p<.0001). 126 Britto, Engle, & Alderman The NECD was designed to enhance the capacity of families and communities to foster the learning and development of children under 6 years of age (preschool or at age of school entry) through enhancing growth and development. The specific aim of the present study was to determine whether there were intervention effects on supportive and stimulating behaviors, father involvement and attitudes towards children's development. Caregiving Behaviors & Attitudes The hypothesis that the project would affect caregiving is supported for attitudes and behaviors related to child learning and stimulation. Compared to baseline, at final assessment intervention mothers are more likely to involve their child in their own daily routines such as housework and agriculture. The child development literature has emphasized the importance of daily routines in learning and development (Boyce, Jensen, James, & Peacock, 1983; Britto, et. al., 2002). By involving children in their own daily routines mothers end up spending more time interacting with their children; an important aspect of caregiving (Bradley, 1995). The results also indicate that intervention mothers reported greater involvement in learning activities at final assessment compared to baseline. For example, intervention mothers reported spending more time in learning activities such as counting, naming, and drawing in the previous day (50%) than control mothers (44%). At baseline, only 25% of both intervention and control mothers reported this behavior. Parents are children's first teachers and their role in their children's development and learning is vital (Richter, 2004; Shonkoff & Phillips, 2000). By enhancing parent's behaviors in these areas the NECD could be considered successful in achieving its aims. Increasing father involvement was a clearly stated aim of the project, and the results provide some support for this aim in terms of changing father's attitudes towards their involvement in children's development (Pruett, 2000). Attitudes towards father involvement beyond traditional aspects of physical caregiving were improved by the project. However changes in actual behaviors were limited. In part greater changes in attitudes and not behaviors could be attributed to the fact that the studied relied on maternal report and mothers tend to do a majority of the caregiving in families for young children (Britto, et. al., 2002). Additionally causing change in father's behaviors is far more difficult in cultures were fathers are not typically involved in caregiving. The most striking changes are noted in caregiving attitudes towards children's learning and development. At final assessment, intervention mothers report greater agreement with their role in promoting children's preparedness for school. Parent's attitudes towards their role in their children's learning and development is important and is seen as an important predictor of children's success in school (Eccles & Harold, 1996; Hoover-Dempsey, & Sandler,1997). By changing parental attitudes towards their role in their child's development and learning the NECD could be considered an important step in achieving its goal of helping families and communities enhance their capacity to improve child outcomes. Study Limitations Notwithstanding the results obtained, limitations of the current study should be noted in interpreting the results. First, the analytical approach used was primarily Early intervention and caregiving 127 descriptive. Multivariate analyses controlling for family and community level confounders were not conducted due to large amounts of missing data on several of the key socio-demographic variables. Given sample size limitations it was not possible to perform analysis of covariance with fixed effects for individual or community. Because communities played a role in selecting the community grants, it is possible that including controls could alter the results obtained. Therefore, as with descriptive results only, these results should be interpreted with caution. Implications for Intervention Programs The key dimensions of effectiveness of successful early intervention programs include timing (age at child enrollment), intensity, duration, comprehensiveness, targeting to the poorest, and quality (Engle et. al.2007; Shonkoff & Phillips, 2000). The program did meet several of these criteria. In terms of timing, the age of children at "enrollment" was young, as children began to receive services at the health center prenatally and at birth, and the educational messages were targeted to parents of young children. Prior results report the greatest effect in the nutrition intervention was in the youngest age group (Alderman, at al., 2003). Second, the program was targeted to the most disadvantaged, where it is likely to have the greatest impact, but there was not a middle SES comparison group. On the dimension of comprehensiveness, the project was relatively strong. It attempted to combine health, nutrition, and development interventions. Although the major focus of the intervention was on growth promotion and health, the effect on caregiving attitudes and behaviours related to both growth and development was notable. However on other dimensions recommended for effectiveness the NECD was less strong. Both the intensity and the duration were limited by several challenges in getting the intervention implemented. First, the population to be covered, but not the funds available, was expanded by request of the government, resulting in lower funds per community than planned. Despite this problem, the communication appeared to be widespread. As noted earlier, in the final survey, almost all (93%) of the project mothers reported hearing of the project, and of those that had, 83% reported face to face contact with someone from the project. Britto et. al. (2005) found that those who reported greater exposure to the project had greater changes in caregiving behaviors. Duration was far less than expected. The intervention was initially implemented by non-governmental organizations, and the process of identifying and training them took longer than originally planned. Therefore, in some areas the evaluation occurred after the program had been in operation less than two years, and some of the interventions had not been implemented as fully as expected. Results may have been stronger with a longer duration, or greater intensity. Quality of the intervention was not assessed. However, given these constraints, the changes in intervention mothers' reported behaviors and attitudes about support for learning are striking. Intervention respondents were much more likely to report being involved in educational activities, and to have positive attitudes toward these activities than control mothers at follow up, whereas there were few differences at baseline. The effects tended to be stronger in behaviors related to learning activities; few changes were seen in play behaviors without this component. Changes noted in father involvement were small, but this is not surprising given difficulties in changing fathers' behaviors without changing broader norms. If a relatively low-duration and intensity intervention can have an impact on parental reports 128 Britto, Engle, & Alderman of these behaviors, stronger interventions have the power to result in far greater changes in a variety of caregiving behaviors. Changes in caregiving behaviors and attitudes must have an impact on a child's development in order for interventions to be justified. Preliminary analyses suggested that the effects for children aged 4-6 years were limited (Alderman et al., 2004). It is possible that greater effects were not seen because the major impact was on younger children who had not yet progressed to the cognitive test sample, given the short duration of the intervention. If family behaviors did change, as reported by mothers, one might expect to see effects on the cognitive development of children aged 4-6 if the children who were under 3 at the time of the intervention were tested at school entry. Changing the cognitive level of children who were already 3-5 at the time of the intervention is more difficult. Brooks-Gunn (2003) in an eloquent title "Do You Believe in Magic? What Can We Expect from Early Childhood Intervention Programs?" reminds us that we must not expect too much from one or two years of a program in early childhood. Yet there is a growing body of evidence that changing early parenting practices can have a subsequent effect on children's development (e.g., Karoly 2005) and such efforts can help us address the loss of human potential of over 200 million children in the developing world (Grantham-McGregor et al., 2007). Adatu, F., Odeke, R., Mugenyi, M., Gargioi, G., McCray, E., Schneider, E., & Maher, D. (2003). Implementation of the DOTS strategy for tuberculosis control in rural Kiboga district, Uganda, offering patients the opetion of treatment supervision in the community, 1998-1999. International Journal of Tuberculosis and Lung Disorder, 7, s63-s71. Alderman, H. (2007). Improving Nutrition through Community Growth Promotion: Longitudinal Study of the Nutrition and Early Child Development Program in Uganda. World Development. 35, 1376-1389. Alderman, H., J., Behrman, R., & Hoddinott, J. (2004). Nutrition, Malnutrition and Economic Growth. In G. López-Casasnovas, B. Rivera and L. Currais (eds.), Health and Economic Growth: Findings and Policy Implications. Cambridge, MA: MIT Press Alderman, H., Britto, P., Engle, P., & Siddiqi, A. (2003). Longitudinal evaluation of Uganda Nutrition and Early Child Development Program. Report to the World Bank, Washington, DC., Unpublished. Alderman, H. and Engle, PL. (in press). The Synergy of Nutrition and ECD Interventions in Africa. In M.Garcia, J. Evans and A. Pence, (eds). ECD in Africa. Washington, DC: World Bank. Alderman, H., Konde-Lule, J., Sebuliba, I., Bundy, D., & Hall, A. (2006). Increased Weight Gain in Preschool Children due to Mass Albendazole Treatment Given During ?Child Health Days? in Uganda: A Cluster Randomized Controlled Trial. British Medical Journal, 333, 122-126. Early intervention and caregiving 129 Alexander, K. L., Entwisle, D. R., & Bedinger, S. D. (1994). When expectations work: Race and socioeconomic differences in school performance. Social Psychology Quarterly, 57(4), 283-299. Ardelt, M., & Eccles, J. S. (2001). Effects of mothers' parental efficacy beliefs and promotive parenting strategies on inner-city youth. Journal of Family Issues, 22, 944–972. Bandura, A. (1997). Self-ef.cacy: The exercise of control. NY: Freeman and Company. Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (1996). Multifaceted impact of self-efficacy beliefs on academic functioning. Child Development, 67, 1206–1222. Barnett, W.S. (1995). Long-term effects of early childhood programs on cognitive and school outcomes. The Future of Children, 5, 25-50. Baume, C.E., Neema, S.T., Kibombo, R., & Cavanero-Verzosa, C. (n.d.). Formative communication research on early childhood development in Uganda. Report prepared for the Nutrition and Early Childhood Development Project, and World Bank. Bornstein, M. H. (Ed.). (2002). Handbook of Parenting (2nd ed.). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Boyce, W. T., Jensen, E. W., James, S. A., and Peacock, J. L. 1983. "The family routines inventory: Theoretical origins." Social Science Medicine 17(4): 193- 200. Bradley, R. H. 1995. "Environment and parenting." In M. Bornstein, ed., Handbook of Parenting, Vol. 2. Hillsdale, N.J.: Erlbaum Bradley, R. H. (1994). The HOME Inventory: Review and reflections. In H. Reese (Ed.), Advances in child development and behavior (pp. 241–288). San Diego, CA: Academic Press. Bradley, R.H., & Caldwell, B.M. (1995). Caregiving and the regulation of child growth and development: Describing proximal aspects of caregiving systems. Developmental Review, 15, 38–85. Bradley, R.H., & Corwyn, R.F. (2005). Caring for children around the world: A view from HOME. International Journal of Behavioral Development, 29, 468-478. Britto, P.R., Brooks-Gunn, J., & Griffin, T. (2006). Story Readers and Story Tellers: Stylistic Differences in Low Income, Young, African-American Mothers' Reading Styles. Reading Research Quarterly. 41, 68-89. Britto, P.R., Engle, P., & Alderman, H. (2005, April).Can caregiving practices be altered: Results from the Uganda Nutrition and Early Child Development Program. Poster presented at the biennial meeting of the Society for Research in Child Development, Atlanta, GA. Britto, P.R., Fuglini, A., & Brooks-Gunn, J. (2002). Reading, rhymes, and routines: American parents and their children. In N. Halfon, K. McLearn, and M. Schuster (eds.), The health and social conditions of young children in American families, (pp. 117-145). New York: Cambridge University Press. Brooks-Gunn, J. (2003). Do you believe in magic?: What we can expect from early childhood intervention programs. Social Policy Report, Society for Research in Child Development, 17 (1). 130 Britto, Engle, & Alderman Brooks-Gunn, J., Klebanov, P. K., & Duncan, G.J. (1996). Ethnic differences in children's intelligence test scores: Role of economic deprivation, home environment, and maternal characteristics. Child Development, 67, 396-408. Cabrera, N. J., Tamis-LaMonda, C. S., Bradley, R. H., Hofferth, S., & Lamb, M. L. (2000). Fatherhood in the twenty-first century. Child Development, 71, 127-136. Custodero, L., Britto, P.R., & Brooks-Gunn, J. (2003). Musical Lives: A collective portrait of American Families. Applied Developmental Psychology, 553-572. Dauber, S. L., Alexander, K. L., Entwisle, D. R. (1996). Tracking and transitions through middle grades: Channeling educational trajectories. Sociology of Education, 1996, 290-307. Dodge, C.P. (1986). Uganda: rehabilitation or redefinition of health services. Social Science Medicine, 7, 755-761. Eccles, S., & Harold, R.D. (1996). Family involvement in children's and adolescent's schooling. In A. Booth & J.F. Dunn (Eds.) Family-school links: How do they affect educational outcomes? (pp. 3-25). Mahwah, NJ: Erlbaum. Engle P.L., Black, M.M., Behrman, J.R., et al. (2007). Strategies to avoid the loss of developmental potential in more than 200 million children in the developing world. Lancet, 369, 229-242. Fuligni, A. S., Han, W., & Brooks-Gunn, J. (2004). The Infant-Toddler HOME in the second and third years of life. Parenting: Science and Practice, 4, 139–159. Grantham-McGregor, S., Cheung, Y.B., Cueto, S., Glewwe, P., Richter, L., Strupp, B. & the International Child Development Steering Group (2007). Developmental potential in the first 5 years for children in developing countries. Lancet, 369, 60-70. Hoover-Dempsey, K. V., & Sandler, H.M. (1997). Why do parents become involved in their children's education? Review of Educational Research, 67, 3–42. Institute Of Public Health (IPH) Makerere University (2003). Re-survey report for the Nutrition and Early Childhood Development Project (NECDP), Uganda Ministry of Health, July 7, 2003. Kampala, Uganda: Draft Report. Karoly, L.A., Killburn, M.R. & Cannon, J.S. (2005). Early childhood interventions: Proven results, future promise. Santa Monica, CA: RAND Corporation. Kinsman, J., Kamali, A., Kanyesigye, E., Kamulegeya, I., Basajja, V., Nakiyingi, J., Schenk, K., & Whitworth, J. (2002). Quantiative process evaluation of a community-based HIV/AIDS behavioural intervention in rural Uganda. Health Education Research, 17, 253-265. Leventhal, T., Selner-O'Hagan, M. B., Brooks-Gunn, J., Bingenheimer, J. B., & Earls, F. (2004). The Homelife Interview from the Project on Human Development in Chicago Neighborhoods: Assessment of parenting and home environment for 3 to 15 year olds. Parenting: Science and Practice, 4, 211-241. McCartney, K., & Phillips, D.A. (Eds.) (2006). Handbook of Early Childhood Development. Oxford: Blackwell Publishing. National Institute of Child Health and Human Development Early Child Care Research Network. (2001). Before Head Start: Income and ethnicity, family characteristics, child care experiences, and child development. Early Education and Development, 12, 545–576. Early intervention and caregiving 131 Pruett, K.D. (2000). Fatherneed: Why Father Care is as Essential as Mother Care to Your Child. The Free Press, New York. Richter, L. (2004). The importance of caregiver-child interactions for the survival and healthy development of young children: A review. World Health Organization. Geneva: Switzerland. Rogoff, B. (2003). The cultural nature of human development. New York: Oxford University Press. Shonkoff, J.P., & Phillips, D.A. (2000). From Neurons to Neighborhoods: The science of early childhood development. Washington, D.C: National Academy Press. Tabachnick, B.G. and Fidell, L.S. (1996). Using Multivariate Statistics. NY: HarperCollins UNSECO (2005) EFA Global Monitoring Report. Paris: UNESCO UNICEF (2006). State of the World's Children. New York: UNICEF. Walker, S.P., Wachs, T.D., Meeks, G. J., and the International Child Development Steering Group Child development (2007). Risk factors for adverse outcomes in developing countries. Lancet, 369, 145- 157. Werner, E. E., Smith, R. S. (2001). Journeys from childhood to midlife: Risk, resilience, and recovery. Ithaca, NY: Cornell University Press. Whiting, B. B., & Edwards, C. P. (1988). Children of different worlds: The formation of social behavior. Cambridge, MA: Harvard University Press Young, M.E. (2002). From early child development to human development: Investing in our children's future. Washington, DC: World Bank.

Source: http://www.childhealthandeducation.ca/articles/documents/4-Paper-Eng-Brito.pdf