|Care and Nutrition - Concepts and Measurement (IFPRI, 1997, 56 p.)|
UNICEF's (1990) original conceptual model of child survival, growth, and development, which identifies the role of care, is presented in Figure 1. In this model, care, household food security, and a healthy environment are the three underlying factors that determine the nutrient intake and health of children, and, in turn, their survival, growth, and development. "Care" refers to practices performed by caregivers that affect nutrient intake, health, and the cognitive and psychosocial development of the child.2
2 Care differs from caring capacity because "capacity" refers to a potential to provide care, but it may not indicate whether the care is provided. If, the caregiver has many constraints to care, she may not be able to put her capacity into practice. A second commonly used term that seems to overlap with care is "infant and young child feeding practices." However, feeding practices, which are usually assessed by interviewing the mother, may reflect overall patterns of behavior rather than specific actions. What people actually do, on the other hand, may or may not be consistent with these general patterns of behavior. For example, a woman may state that she introduced complementary feeding to a child at four months of age and forget the small bites offered to a curious child at three months because they did not seem to be significant events and were not encoded in her memory as foods offered. If she thinks that her overall decision was to introduce complementary food at four months, the small bites earlier may be inconsistent with her overall pattern and therefore not remembered. Thus the reported infant feeding practice represents a simplification of practices consistent with belief, but not necessarily an accurate reflection of day-to-day practices. Both practices and behavior are important to evaluate, but the difference between them should be recognized.
This model of care can be expanded in two directions. First, it should emphasize that effective care practices require time and other resources, and second, it should underscore that the child's behavior or characteristics play a role in determining care.
The Extended UNICEF Model of Care
In order to perform care practices, the caregiver needs sufficient education, time, and support. The provision of these resources by family or society can be considered care for the caregiver. In Figure 2, the UNICEF model is adapted to incorporate care to the caregiver. Six major categories of resources that caregivers need can be identified from the literature: (1) education, knowledge, and beliefs; (2) health and good nutritional status; (3) mental health, lack of stress, and self-confidence; (4) autonomy, control of resources, and control of intrahousehold allocation; (5) reasonable workloads and adequate time available; and (6) social support from family members and the community. These six are the human, economic, and organizational resources identified in the UNICEF model, defined at family and community levels (Jonsson 1995).
Education, knowledge, and beliefs represent the capacity of the caregivers to provide appropriate care. The physical and mental health (including self-confidence and lack of stress and depression) of the caregiver represent individual factors that facilitate the translation of capacity to behavior. Finally, autonomy, workload, and social support are facilitating conditions in the family and community. Some of these resource categories have been investigated extensively, whereas others have been investigated primarily in developed countries or await further investigation.
The Transactional Model of Care
The extended UNICEF model of child care is a useful framework for assessing the capacity and ability of the caregiver and family to provide care. However, a model of child care should assess not only the caregiver's behavior, but also the behavior of the child and the characteristics of the child's environment. All three of these factors play a significant role in the eventual nutritional status of the child (Black et al. 1994).
For the past 25 years, psychologists have documented the significant role that children play in determining the care that they receive (see, for example, Bell 1971). Differences between children, such as endowed healthiness, perceived vulnerability, perceived weight, and even physical attractiveness, affect the practices of their caregivers3 The transactional model of care argues that the results or effects of a child's endowments are a function of a long series of mutual interactions or transactions between the developing child and the care-giver, and that these interactions are constantly changing with the changing developmental status of the child (Sameroff 1989). At the heart of the process is the relationship between the child and the caregiver (or care-givers). This affective, or emotional, relationship is a unique and life-long bond between two humans, called an attachment (Ainsworth et al. 1978). Problems in this emotional relationship can contribute to child malnutrition or ill health or may result in attachment problems. For example, Valenzuela (1990) found that in Chile, children who were undernourished were far less likely to be securely attached to a caregiver, though this association could not be interpreted as causality.
3 See Engle and Riccuiti 1995 for a summary of this argument.
Figure I - The UNICEF conceptual model
Source: UNICEF 1990.
Figure 2 - The extended model of care
The healthy development of a child has been found to depend on the development of a secure attachment or a close bond with at least one caregiver during infancy, from whom the child received abundant positive attention (Werner 1993). Attachments have been divided into those that are secure (about two-thirds) and those characterized by ambivalence and the child's avoidance of the caregiver in middle-class samples (Ainsworth et al. 1978). Attachment can be assessed in a standardized situation and has been measured in many different cultures, although there are some questions about the validity of the measurement of attachment across cultures (Becker and Becker 1994).
A critical aspect of quality of care seems to be responsiveness to the child's cues, verbalizations, signals, and so forth (see, for example, Bronstein 1991). Responsiveness does not mean that the caregiver always gives the child what is requested, but that the caregiver's response takes the child's needs and developmental level into account. Not acceding to inappropriate demands by active and well-nourished children is an important part of a caregiver's responsiveness. Usually, a positive emotional (affective) relationship between caregiver and child will be reflected in warm and responsive caregiving practices. However, the lethargic or unresponsive child will have a harder time stimulating a caregiver's responsiveness.
The extended UNICEF model can be adapted to include the relationship between child and caregiver (Figure 3). This figure expands the central part of the UNICEF model relating care, nutrient intake, health, and child growth and cognitive development. Eleven specific arrows have been drawn to illustrate the various ways in which the affective relationship between caregivers and child and the resulting care practices can influence the child's growth, cognitive and psycho-social development, dietary intake, and health status, and how child growth and development may influence care and the affective relationship.
Arrows 1 through 4 represent well-known linkages. Arrow 1 links growth and cognitive development. This relationship has been demonstrated in numerous studies, including three of the Collaborative Research Support Program (CRSP) Nutrition studies, funded by the U.S. Agency for International Development (USAID) (Kirksey et al. 1992; Alien et al. 1992; Neumann, Bwibo, and Sigman 1992), although the reasons for the linkages are not entirely known. The initial hypothesis of a linkage between energy and protein intake and brain growth has not received unequivocal support (Pollitt et al. 1993; Engle et al. 1993), although increasing interest in micronutrient deficiencies may again lead to a brain model for explaining these effects. Lozoff, Jimenez, and Wolf (1991) offer one example.
Arrow 2 suggests that the caregiver-child relationship is important for the child's cognitive and psycho-social development. Many psychological studies support this linkage.4 Arrow 3, the link between dietary intake and growth, is well known, as is Arrow 4, the link between morbidity and slower growth.
4 See, for example, Rutter 1990; see Engle, Castle, and Menon 1996 for a summary.
Arrows 5 through 11 represent the effects of interaction between the child and the caregiver. Arrow 5 suggests that a child who has a higher level of cognitive development will be better able to build a positive affective relationship with the caregiver. Although fewer data exist to support this linkage, recent findings from the Nutrition CRSP studies illustrate the associations between child vocalizations and type of interaction with the caregiver (Chavez et al. 1987; Sigman et al. 1989).
Arrow 6 suggests that the nature of the affective relationship with the caregiver can influence child growth. Early studies of infants raised in orphanages with no consistent caregiver found significant improvement in growth and cognitive development of children when they were assigned to a particular caregiver who routinely provided care and stimulation (Dennis 1973). More recently, a series of studies by Field and colleagues with preterm infants indicate that firm massage on a daily basis will result in increased weight gain even without additional foods (Field 1992, 1993; Field et al. 1986). This effect, which has also been illustrated with rat pups, may occur as a result of stimulating the growth hormone (Schanberg and Field 1988). Although the variable manipulated in this case was not attachment, touching and stroking of infants is usually a component of a positive and responsive relationship.
Figure 3 - The transactional model of care
Arrow 7 suggests that the child's growth may be associated with the affective relationship and, therefore, may affect the care received. Some data suggest that better nourished, larger children may receive more care (Arya 1989), and that under some conditions, a poorly nourished child may be assumed to have no will to live and be allowed to die (Scheper-Hughes 1992). A study in Mexico found that mothers interacted more with better nourished, larger children than with smaller children (Alien et al. 1992). In addition, the lack of responsiveness of low-birth-weight children has been associated with poorer feeding practices by mothers (Barnard et al. 1989).
Arrow 8 suggests that the amount that a child eats can play a role in the relationship of the caregiver to the child. Caregivers may be particularly troubled by a child who refuses to eat; feeding difficulties are one of the most common behavioral disturbances of young children reported to pediatricians in industrialized countries (Sanders et al. 1993). In the United States, 24 percent of two-year-olds and 19 percent of three-year-olds were reported by parents as having feeding problems (Beautrais, Fergusson, and Shannon 1982). A number of studies have compared children with non-organic failure to thrive (NOFTT) with normally growing children; many such studies have observed inadequate interactions between parent and child among the NOFFT group (Black et al. 1994). The causality is difficult to untangle, although one study did find a higher incidence of oral-motor difficulties in NOFTT children, suggesting that the child's feeding problems may precipitate poor child-caregiver interactions (Mathisen et al. 1989). The much higher rate of NOFTT among low-birth-weight children (Kelleher et al. 1993) also suggests that a child's poor growth may contribute to a problematic caregiver-child relationship or to failure to thrive.
Arrow 9 proposes that a positive affective relationship leads to increased dietary intake. In the failure-to-thrive literature, observational studies suggest that mothers of NOFTT children in affluent societies tend to be less attentive to their children and interact and verbalize less, and their children have shorter feeding episodes and ingest less food than well-nourished children (Black et al. 1994; Heffer and Kelley 1994).
Arrow 10 indicates that the health of the child may influence the nature of the affective relationship, leading to either an increase in attention and caring from the caregiver or, at some point, a reduction in investment if the child is perceived as having little chance of survival (Scheper-Hughes 1992; Cassidy 1987).
Finally, the quality of the relationship of the caregiver to the child may influence the health status of the child by affecting health care treatment and the health care sought (Arrow 11). Although direct evidence for this relationship is lacking, differences in health-seeking behavior by gender of the child frequently reported in the Indian subcontinent may be examples of caregiver preference influencing health care treatment. That is, treatment is more likely to be sought for boys than for girls in that region. However, without further evidence, it is impossible to determine whether the reduction in health care seeking is a function of the quality of the relationship or of maternal and family strategies of investment in child care (Alderman and Gertler 1996).
These arrows illustrate the central role of the affective context of caregiving, but much remains to be learned about strategies for changing this context to enhance child survival, growth, and development.