This is the old United Nations University website. Visit the new site at

Previous Page Table of Contents Next Page

Overview of early child-care and education programmes and Jamaican case studies

Intervention programmes in the United States
Programmes for children over two years of age: Centre based, home based, or both
Long-term follow-up after intervention
Intervention programmes in developing countries
Summary and conclusions

Christine Powell

The author is affiliated with the Tropical Metabolism Research Unit in the University of the West Indies, Mona, Kingston, Jamaica.


The extent to which early childhood programmes can produce long-term benefits to children’s cognitive and social-emotional development continues to be a major concern of policy makers. This paper examines some of the model intervention programmes for children under four years of age that have been carried out in the United States and some developing countries, with emphasis on our experience in Jamaica. In general, programme participants have shown concurrent gains in IQ during the intervention. These have been sustained into the early school years, after which there is a tendency for the gains to decline. However, some programmes have reported persistence of IQ gains to the age of 12 years. Long-term gains in educational progress include fewer placements in special education classes and less grade retention. Programmes of greater duration and intensity were more likely to be successful.


Over the last four decades, there have been several studies designed to evaluate the impact of early child-care and education programmes on the cognitive and social-emotional development of children from poor circumstances. Although these programmes have been instituted in many countries, the United States provides the largest number of evaluations carried out under experimental conditions, which can give some estimate of the benefits to be derived. I will give a brief overview of some of the more well-designed programmes from the United States for children under four years of age, followed by some programmes from developing countries. I will conclude with some Jamaican case studies.

In general, programmes may be categorized according to who was targeted - the parent, the child, or both - or whether the intervention was home or centre based. Centre-based programmes are usually conducted outside the home in day-care centres or specially designed pre-school centres. These programmes are concerned mainly with directly affecting the child rather than the family, although in some cases instructions are also provided for the mothers in centres. A few have been entirely parent focused, in that parents attend regular meetings at a centre.

Home-based programmes are usually aimed at improving the child’s developmental status by enhancing the child-rearing and child-care environment in the home. This is generally done through home visits intended to influence the child’s primary caregiver. The target is therefore one or both parents and the child. It was expected that working with parents would help to combat post-programme erosion effects by providing a mechanism for stimulation to continue after the programme ended. There was also the possibility that benefits would spread to other children within the family and possibly also to neighbours. It was also expected that mothers would derive benefits for themselves, and many programmes included activities to help them develop a more positive self-concept, feelings of competence, and skills training.

Some of the better-evaluated programmes from the United States, which had either an experimental or a quasi-experimental design, are summarized in tables 1 to 3. They were also chosen to reflect the variety of approaches that have been tried.

Intervention programmes in the United States

Programmes for children under two years of age: predominantly centre based
Programmes for children under two years of age: predominantly home based

Programmes for children under two years of age: predominantly centre based

Some of the centre-based programmes involved intensive interventions for the children that required costly personnel and material resources; for example, the Milwaukee study [1] involved extensive provisions for the child and family for six years. The children were enrolled as early as six weeks to three months of age and were randomly assigned to experimental or control groups. They spent full days in the centre, where the staff-to-child ratios were as low as 1:2 for the babies and increased as the children got older. Family intervention that provided job training and social and remedial education for the mothers was also included. Some of the largest gains in IQ have been reported from this programme; for example, differences as high as 32 IQ points between the children in the experimental and control groups were reported at three years of age. At 10 years of age, the children in the experimental group had superior school achievement and a mean IQ 20 points higher than that of the control children. There has been some controversy over the persistence of these gains, and the programme has been criticized as unrealistic, impractical, and not reproducible on a large scale.

A similar intensive centre-based programme, the Carolina Abecedarian Project [2], was followed by a school-age intervention in which a resource teacher visited the homes and provided the parents with individualized sets of educational activities for the children. There were four groups: centre alone, centre plus three years home instructions, home instruction alone, and no intervention. At follow-up at 12 years of age, the children who had participated in the centre-based preschool programme, alone or in combination with the school-age programme, had significantly higher IQ and school achievement scores than the controls. The advantage of the additional home intervention was small. Project CARE [3] extended the Abecedarian Project concept to include a family educational intervention. This was influenced by the thinking that a family-support intervention would be effective in changing the home environment and possibly have a larger impact than the more child-focused centre-based programme. Three groups were compared: centre-based programme plus family education delivered through home visits, family education alone, and no intervention. The centre plus family-education group had significantly higher scores than the other two groups on the Bayley Mental Development Index up to age 18 months and on the Stanford-Binet test up to 36 months. At 48 months the centre group had higher scores than the family-education group but not the controls. Thus, family education alone did not have any benefit on the children’s cognitive development.

The Parent Child Development Centers [4] in three states (Louisiana, Texas, and Alabama) focused predominantly on the mothers, who met in groups at a centre to discuss activities related to the children’s development. Sessions on family relationships were also included to help the mothers function more effectively within the family. The expectation was that mothers would retain benefits and help to sustain those seen in the children. There were variations among the three programmes in time spent in the centre (table 1). The Houston, Texas, programme included home visits in the first year, complemented by four weekend family workshops. In the second year, mothers and children attended the centre four times a week. In the Birmingham, Alabama, group the mothers assisted with the care of their own children in the centre and gradually took over responsibility as the children got older. Participants in all three programmes were randomly assigned to intervention or control. Benefits to the children during the intervention ranged from 6 to 8 developmental quotient (DQ) points, and significant benefits were detected one year after the intervention in the groups in Birmingham and New Orleans, Louisiana. In Houston advantages in the achievement levels in grades 2 to 5 were reported, but no benefit was found to IQ at 36 months of age.

Instructions that were meant to improve the cognition and verbal development of the child were sometimes given directly to the mother in the absence of the child. Karnes and colleagues [5] combined such activities with home visits to observe how the mothers were implementing the programme. A difference of 16 IQ points on the Stanford-Binet test between index children and controls was observed after 15 months. This improvement is comparable to that seen in other studies in which trained professionals were used. Although the inputs were considerable, including the availability of play materials and monthly home visits as well as weekly meetings, it is impressive that these gains were made with such limited contact with the child. However, the children were not randomized to groups, and therefore caution must be used in interpreting the results. The economic advantage of using mothers to carry out the intervention could have far-reaching implications in the extension of such services.

High-risk children who were also socio-economically disadvantaged have also shown significant gains from early intervention. The Infant Health Department Program [6] was a multisite intervention for low-birthweight children ($ 2,500 g). The programme, which lasted from birth to three years of age, included weekly home visits in the first year and fortnightly thereafter, educational centre-based care from 12 months, and parent support groups. By three years of age, children who weighed between 2,000 and 2,500 g at birth had an average IQ 13.2 points higher than the controls; the gains among children with lower birthweights were more modest.

Programmes for children under two years of age: predominantly home based

Summaries of four predominantly home-based programmes that were initiated in children under 24 months of age are shown in table 2. Home visits were conducted on a weekly or fortnightly basis but did not necessarily involve activities with the child. Home visitors ranged from professional teachers [7] to peers of the mothers [8]. The Florida Parent Education Programme [8] is an example in which paraprofessionals or peers of the mothers were used to carry out the intervention. This programme had several groups, for which the length of intervention varied from one to three years. In addition to home visits, the children also spent two mornings per week in a centre during the third year of the programme. Starting at the age of 3 years, the children who had received two or three consecutive years of intervention had significantly higher scores on IQ tests, and the differences were still significant at 10 years of age. Benefits in school achievement and a reduction in placement in special-education classes were also related to the intervention.

TABLE 1. Intervention studies in the United States with children from birth to two years of age: predominantly centre based

Program & ref.

Initial age

Sample & assignment


Child outcome

Milwaukee Project [1]

3 mo

Low-SES mothers with low Iqs
Random assignment to experimental or control group until age 6 yr

Child spent 5 da/wk in centre

Experimental group significantly better than controls on Gesell at 18 mo, WISC up to 10 yr, & school achievement at 10 yr

Carolina Abecedarian [2]

6 wk-3 mo

Low SES Random assignment to experimental or control group

1. Control
2. Centre-based programme to age 5 yr
3. Centre-based programme to age 5 yr+ home instruction to age 8 yr
4. Home instruction to age 5-8 yr

Experimental preschool group significantly better than controls on Bayley to 18 mo, Stanford-Binet to 48 mo, & WPPSI at age 5 yr
Experimental pre school with or without school-age intervention significantly better than school-age intervention alone or no intervention on IQ tests at age 12 yr

Project CARE [3]

6 wk-3 mo

Random assignment to 2 experimental groups or control group for 5 yr

1. Child attended centre + family education through home visits
2. Family education only
3. Control

Centre + family education group significantly better than other 2 groups on Bayley at 12 & 18 mo & on Stanford-Binet at 24 & 36 mo
Better than family education only group but not controls at 48 mo

PCDC, New Orleans, LA, USA [4]

2 mo

Random assignment to experimental or control group for 3 yr

Parent meetings 2 mornings/wk & monthly group meetings

No significant difference between experimental & control groups on Bayley & Uzgiris & Hunt at 22 mo
Stanford-Binet scores significantly higher in experimental group at 36 mo & 1 yr later

PCDC, Houston, TX, USA [4]

12 mo-3 yr

Random assignment to experimental or control group for 2 yr

Weekly home visits in yr 1
Child spent 12 h/wk in centre in yr 2

Experimental group significantly better than controls on Bayley at 24 mo
No difference on Stanford-Binet at 36 mo or school achieve ment in grades 2-5

PCDC, Birmingham, AL, USA [4]

3-5 mo

Random assignment to experimental or control group for 3 yr

Child attended centre with mother 3.5 da/wk in yr 1,4.5 da/wk in yr 2, 5 da/wk in yr 3

No difference between experimental & control groups on Bayley at 22 mo
Stanford-Binet scores significantly higher in experimental group at 36 mo & 1 yr later

Multisite IHDP group [6]


Low-birthweight & premature (<2,500 g& 37 wk)
Random assignment to experimental or control group for 3 yr

Weekly home visits in yr 1, bi-weekly in yr 2 & 3
Child attended centre from 12 mo to 3 yr

Experimental group significantly better than controls on Stanford-Binet at 36 mo
Children weighing 2,000- 2,500 g at birth had greatest gains

Educational Intervention at Home [5]

12-24 mo

Matched controls
Duration 7 mo in yr 1& 8 mo in yr 2

Weekly group meetings for mothers + monthly home visits

Experimental group significantly better than controls on Stanford-Binet & ITPA at end
No follow-up reported

Abbreviations: IQ, intelligence quotient; IHDP, Infant Health and Development Program; ITPA, Illinois Test of Psycholinguistic Abilities; PCDC, Parent Child Development Center; SES, socio-economic status; WISC, Wechsler Intelligence Scale for Children; WPPSI, Wechsler Preschool and Primary Scales of Intelligence.

Professional teachers conducted weekly home visits of 60 to 90 minutes duration over 16 months in the Ypsilanti-Carnegie Infant Education Project [7]. Although some differences were reported during the intervention period in favour of the experimental children, no significant differences were found in the IQs of the children at the end or one and five years later. In the Nashville Programme, children were randomly assigned to extensive home visiting, materials only, or controls for nine months [9]. The only difference found among the groups was a significant increase in receptive language scores, a test of language interaction between mother and child in the extensive home-visiting group. Gutelius and colleagues [10] conducted an intervention provided to the family by paediatricians and nurses who used a mobile unit to provide prenatal counselling and well-baby care, supplemented by home visits from the nurses for cognitive stimulation. The programme was started while the mothers were pregnant. They were counselled on how to provide appropriate sensory, motor, and language stimulation for the child. An average of 21 health and 24 stimulation visits were carried out over the three-year period. By 12 months, significant differences had emerged, and by 36 months, the difference of 8 points on the Stanford-Binet test was highly significant. These benefits are important, considering the small amount of input that was spread over three years. This is an example of integrating a stimulation programme through the health services.

TABLE 2. Intervention studies in the United States with children from birth to two years of age: predominantly home based

Program & ret.

Initial age

Sample & assignment


Child outcome

Florida Parent Education [8]

3 mo

“Indigent” on obstetric records
Random assignment to experimental or control group for 3 yr

Weekly home visits for 3 yr & 2 mornings in centre during 3rd yr

Experimental group not different from controls at 12 & 24 mo
Experimental group significantly better than controls on Stanford-Binet at 3-6 yr, WISC at 10 yr, & school achievement at 10 yr

Ypsilanti- Carnegie Infant Education [7]

3, 7, & 11 mo

Random assignment to experimental or control group for 16 mo

Weekly home visits

Experimental group not different from controls on Bayley or Stanford-Binet at end or 1 yr later

Family-Oriented Home Visiting[9]

17-24 mo

Random assignment to experimental or control group for 9 mo

Weekly home visits

Experimental group not different from controls on Stanford- Binet at end or 2 yr later, but receptive language significantly different at end

Mobile Unit for Child Health Supervision [10]

7th mo of pregnancy

Unmarried teen-aged mothers
Random assignment to experimental or control group for 3 yr

21 health-care visits &24 stimulation visits

Experimental group significantly better than controls on Bayley at 12&24 mo & Stanford-Binet at 36 mo

Abbreviations: SES, socio-economic status; WISC, Weschler Intelligence Scale for Children.

The results from these programmes were mixed, in that benefits to the children’s cognitive development were seen in only two of the programmes, with the largest improvements coming from the Florida Parent Education Programme, which was conducted over a three-year period. The benefits were generally smaller than those seen in the predominantly centre-based programmes and raised questions concerning the efficacy of home-based programmes in that setting.

Programmes for children over two years of age: Centre based, home based, or both

Summaries of three programmes that began after two years of age are shown in table 3. The Early Training Project [11,12] consisted of an intensive 10-week centre-based programme during the summer, followed by weekly home visits for the remaining nine months of the year. One group of children received three years of intervention, another received two years, and a third served as a control. There was an increase in the children’s IQ scores during the intervention, but these benefits gradually declined once the intervention had ceased.

In the Perry Preschool Project [13,14], children from low-income families with IQs ranging from 70 to 85 were recruited in cohorts over a four-year period. They were randomly assigned to experimental or control groups after stratification for IQ, sex, and socio-economic status. Children in the experimental group were taught from a cognitively oriented curriculum in a preschool centre for 12.5 hours per week and were visited at home with their mothers for 90 minutes each week. The first cohort spent 30 weeks in the programme, and subsequent cohorts spent 60 weeks.

This project provided one of the longest and most intensive follow-ups of such programmes and gives insight into not only cognitive benefits but also socialization and adult economic success. The programme effects have also been analysed in terms of the costs and benefits. The programme resulted in benefits to the children’s IQs during the intervention, and although they declined somewhat after intervention, they were still significantly higher than those of controls up to 2 years later. Significant differences were found in school achievement at 9 and 14 years of age. In addition to better school adjustment, as reflected by fewer placements in special education and grade retention, the experimental group also had fewer episodes of delinquency and adult criminality and greater adult economic success.

TABLE 3. Intervention studies in the United States with children two to four years of age: centre based, home based, or both

Program & ref.

Initial age

Sample & assignment


Child outcome

Early Training Project [11,12]

3.5-4.5 yr

Random assignment to experimental or control group for 1 or 2 yr

Summer pre-school for 10 wk
Weekly home visits for rest of year

Experimental group significantly better than controls on WISC at end
3 yr later no difference between groups

High/Scope Perry Preschool Project [13,14]

3 or 4 yr

Random assignment to experimental or control group, stratified by IQ
Duration 30 or 60 wk

Weekly home visits
Child attended centre for 12.5 h/wk

Experimental group significantly better than controls on Stanford-Binet after 1& 2 yr & after 2 yr follow-up
School achievement better at age 14 yr

Verbal Interaction Project [15-17]

2 or 3 yr

1. Matched 3 groups for 1 yr, 3 groups for 2 yr, 3 control groups
2. Pairs of experimental & control groups

Bi-weekly home visits

Marked IQ difference at end of study between experimental & control groups
School achievement & Stanford-Binet significantly better in experimental groups in grade 3

Abbreviations: IQ, intelligence quotient; SES, socio-economic status; WISC, Weschler Intelligence Scale for Children.

Levenstein and colleagues [15, 16] conducted a series of home-based interventions for children aged two to three years. Mothers and children were visited by toy demonstrators twice a week for half-hour sessions. The toy demonstrators worked from a structured curriculum and involved mother and child together in play. Children who received two years of visiting had larger benefits than those receiving one year. The mean difference in IQ between the two-year group and the controls was 13 points. At eight and nine years of age, the children who had received intervention showed benefits in school achievement and in expected grade placement. However, the differences in IQ were no longer significant [17].

Long-term follow-up after intervention

One of the major concerns regarding the effectiveness of pre-school intervention programmes is whether their benefits persist. The Consortium for Longitudinal Studies [18] provided some evidence of long-term effects of early education programmes in a follow-up study of children who had participated in 11 projects that provided special educational programmes for children from families of low socio-economic status. The children were 8 to 18 years of age and were assessed by psychological and educational measures. Their school records were collected, and they and their parents were interviewed. The studies varied in a number of characteristics, such as the age of the child at the start of the project, the extent of parental involvement, curriculum models, types of programme delivery, and length and intensity of the intervention. The studies were analysed separately and together.

The pooled results showed that children who had participated in pre-school programmes had better school adjustment as indicated by lower rates of grade retention and referral to special-education classes. The programme participants also had a more positive self-concept, more achievement-related reasons for being proud of themselves, and higher rates of school completion and employment. The impact on the families was also important. Programme parents showed higher aspirations and expectations for their children and indicated more satisfaction with their children’s school progress than the parents of control children.

The results of school achievement testing were somewhat disappointing, in that only modest gains were seen. It might have been expected that the improved school adjustment would have been reflected in school achievement. The IQs of the programme children were significantly higher than those of the control children during the period of intervention, with an overall difference of 7.4 points at the end after background variables were controlled for. After this there was a gradual decline, although the effect was still significant up to three or four years later.

The mechanism proposed was that children who had participated in pre-school programmes had gained skills such as the ability to follow instructions, task perseverance, and sustained, focused attention, and that these, together with the cognitive gains, had helped them to adjust to the demands of the school system. These positive attitudes and behaviours had in turn elicited positive responses and reinforcement from teachers and parents, thereby helping the children to develop a greater sense of competence and higher aspirations for success. Although the participants in the interventions did better than the control children, it is important to note that they were still not functioning at the level of middle-class children.

Barnett [19] reviewed 36 early childhood programmes in the United States, including 15 model demonstration programmes and 21 large-scale public programmes. He confirmed the earlier findings of the Consortium, in that the programmes resulted in improved performance on IQ tests and that these remained at least until they entered school, after which there was a tendency for these gains to decline. Other types of long-term impacts on subsequent school progress, such as lower rates of grade retention and placement in special classes, were also seen. The studies with the more rigorous study designs found the greatest effects, which suggests that design flaws and measurement errors diminished some of the actual programme effects. Ramey and Ramey [20] concurred that well-designed programmes with intensive interventions could produce long-term benefits. Boocock [21] reviewed studies from other developed countries and also found similar positive outcomes.

Intervention programmes in developing countries

Studies in various countries
Jamaican studies

Studies in various countries

Although pre-school programmes and child-care services have been offered in many developing countries, relatively few have been rigorously evaluated. Two experimental studies that were carried out in response to the growing concern over the effects of malnutrition on mental development in Cali [22] and Bogota [23] in Colombia showed long-term benefits. They included nutritional supplementation and health care, alone or in combination with educational intervention. In the Cali study, which was centre based, the combined interventions had substantial benefits to the children’s development that were proportional to the duration of the intervention. The children receiving intervention made more progress in the first three grades in school and had higher IQ scores at nine years of age [24]. The Bogota study was conducted with high-risk families. The intervention began with pregnant women, who received nutritional supplementation. The children received supplementation for three years, either alone or in combination with maternal education. All interventions benefited the children, and the combined interventions produced the largest gains. No detailed follow-up of the children’s development has been reported. A brief report indicated that the supplemented group performed better on tests of school readiness at seven years of age [Mora JO, Super CH, Herrera MG, personal communication, 1991]. The main conclusions from these studies are that some of the deficits in intellectual functioning that accompany poverty and malnutrition can be prevented or ameliorated through improvements in diet, learning experiences, and health surveillance.

There are also evaluations of existing service programmes that have been reviewed by Halpern and Myers [25] and Myers [26]. These include programmes from Brazil, Chile, Colombia, and Peru that have used a variety of approaches, such as parent education and day-care centres for the children. In general, modest gains in children’s cognitive development and in mothers’ knowledge were reported. However, few studies had comparison populations, and there was little attempt at long-term follow-up.

The Turkish Comprehensive Early Enrichment Programme [27, 28] evaluated the impact of three different types of pre-school care: educational pre-school centres, custodial day-care centres, and home-based day care. Half of the mothers in each group received an additional training component on children’s development, and the results were compared with those without maternal training.

The children in the educational centre performed better on tests of IQ and cognitive functions and achievement in mathematics and Turkish, and had higher school grades up to grade 3, than children in the other groups. Maternal training also increased the children’s IQ scores, but there was a tendency for maternal training to benefit the children in the custodial and home programmes more than those in educational centres. The problem with interpreting these results is that children from educational centres came from better homes. However, there appears to have been a clear effect of the maternal training programme.

The programme of Integrated Child Development Services (ICDS) [29] in India was designed to provide supplementary nutrition, immunization, health care, health and nutrition education, and non-formal preschool education to children under six years of age and education to women of child-bearing age, especially pregnant and lactating women. These extensive services, covering several million children and mothers, are delivered by paraprofessionals in group settings.

Chaturvedi looked at the psychological and social development of the children in one of the original experimental areas [29]. A group of 214 project children was compared with 208 children matched for parental education and occupation, socio-economic status, and household education. The children had been exposed to the ICDS for a continuous six-year period and were six to eight years old at the time of evaluation. They were significantly better than non-ICDS children in school attendance, academic performance in school examinations, and general behaviour in school, as rated by the teachers. At ages six and seven, they had significantly higher IQs. No long-term follow-up was reported.

Jamaican studies

We carried out five studies with various groups of deprived children in Jamaica. The typical pattern of development of children is precocity in the first year of life as compared with test norms and then a gradual decline thereafter [30]. By four years of age, they could be up to 18 months behind their more advantaged peers [31] whose development is comparable to, or above, the test norms [32].

Four of the studies used the home-visiting approach in which a trained visitor worked with the mother and child in the home. The approach was based on the assumption that by working with mothers and children we could bring about changes in the mothers’ child-rearing practices that would make them more effective teachers of their children. The visits were aimed at enhancing mother-child interactions and improving the self-esteem of both. In working with the mothers, we also hoped that the benefits would last longer and spread to other children in the home and possibly to neighbourhood children as well. This model may be less expensive than centre-based programmes and therefore more feasible, particularly for developing countries.

The home visits were carried out by trained community health aides, except in the first study, in which trained nurses were used. A visit would last for approximately one hour, during which the visitor would explain activities to the mother and demonstrate play techniques by playing with the child. We developed a curriculum based on Piagetian concepts, such as search for hidden objects, in children under 24 months of age.

For those older than two years, concepts such as colour, shape, size, quantity, position, and motion were used. In addition, activities that facilitated language development, fine and gross motor skills, and problem solving were included. Simple instructions were written for each toy and activity. Toys were taken and left until the next visit, when they were exchanged for new ones [33]. Mothers were encouraged to play with their children between visits, praise them, and give them positive reinforcement. A strong emphasis was placed on verbal interaction. The relationship between the visitor and the mother was friendly rather than authoritarian.

To maintain the quality of the visits and to ensure that the activities conducted were appropriate for the child’s ability, the visitors attended regular workshops with a supervisor. The visitors kept records of the visits, which were discussed and used to plan the next visit. Four to five visits were done per day, and the supervisor accompanied the visitor at least twice per month and completed an evaluation checklist [34].

Intervention with community children

In the first intervention project, 22 children aged 34 to 40 months were visited at home weekly for one hour. This project was fairly expensive, since the visitors were nurses and the toys were bought. After eight months, the visited group showed an increase of 13 DQ points as compared with a group of children from a neighbouring community who were not visited [34]. In subsequent programmes, we developed a range of toys made from materials commonly available in the homes, such as plastic bottles, margarine containers, and scraps of cloth. This helped to reduce costs and had the additional benefit of demonstrating to the mothers toys they could make from materials available to them.

The next community intervention was run from a government health clinic, and the intervention was started at a much younger age, as the focus was to prevent the decline in development that is frequently seen in young children from poor homes. The aim was to determine whether the programme could be delivered through the existing primary health-care services supervised by a midwife and using community health aides as visitors. We also wanted to determine the relative effectiveness of different frequencies of visiting schedules. The study had two phases.

In the first phase, three groups of children aged 6 to 30 months assigned by neighbourhood were chosen. One group of 49 children was visited fortnightly, and another group of 45 children was visited monthly for two years. The third group of 45 children served as controls. The group visited fortnightly showed a modest but significant improvement in development as compared with the group visited monthly and the controls. The DQs of children in the group visited monthly were not significantly different from those of controls. How- ever, they had significantly higher scores on the Peabody Picture Vocabulary Test than the controls, as did the group visited fortnightly. This was the only benefit to the group visited monthly.

The effect of fortnightly visiting was less than anticipated on the basis of previous experience with weekly visiting. It was not clear whether this was due to less skilled visitors or to the reduced frequency of visits. A weekly visited group was therefore introduced. Fifty-eight children aged 16 to 30 months from the same neighbourhoods were randomly assigned to intervention or control. After one year of intervention, the mean DQ of the group receiving intervention was 11 points higher than that of the controls. The DQs of the children in the group visited weekly were significantly higher than those of children of similar age in the groups visited fortnightly and monthly during the first phase [35]. The benefits therefore increased as the frequency of visiting increased, giving a dose-response relationship (fig- 1).

FIG. 1. Mean developmental quotients (DQ) of children aged 16-30 months in experiments 1 and 2 [35]

FIG, 2. Mean standard IQ scores of severely malnourished children who received intervention compared with scores of malnourished and adequately nourished children who did not receive intervention followed up 14 years after leaving the hospital [38]. Children were assessed first by the Griffiths Test up to 4 years after leaving the hospital, then by the Stanford-Binet Test, and at 14 years by the Weschler Intelligence Scale for Children

These results showed that the development of poor urban children could be improved with psychosocial stimulation. Improvements varied in proportion to the frequency of visiting, and weekly visiting made the most substantial improvements to the children’s development. It was also shown that this type of intervention could be delivered from a health centre. However, the level of supervision was much higher than might be possible on a day-to-day basis.

Intervention with severely malnourished children

A similar home-visiting approach was successfully used with malnourished children. In one study, severely malnourished children aged 6 to 24 months were played with daily while in the hospital and then visited weekly at home for two years and fortnightly for a third year. They were compared with a group of severely malnourished children who were admitted to the same hospital the previous year and received standard care without a play intervention, and with a group of adequately nourished children who had been hospitalized for other reasons [36]. They were followed for 14 years after leaving the hospital. Both malnourished groups had similar low levels of development on admission to the hospital. The malnourished children who received intervention showed marked improvement and actually caught up to the controls during the intervention. However, the malnourished children who did not receive intervention continued to show a large deficit relative to the controls throughout the study (fig. 2). At the 14-year follow-up, the group receiving intervention showed benefits in the Weschler Intelligence Scale for Children full and verbal scales and in school achievement scores as compared with the control group [37].

Combined intervention with nutritional supplementation and psychosocial stimulation

After our studies of the effects of providing child development activities to severely malnourished and poor community children, it was important to determine whether additional benefits would be gained by providing nutritional supplementation as well as child development activities and whether the benefits would be interactive or additive. In a fourth study [38], we worked with stunted children, since up to 40% of children under five years of age in developing countries are stunted. One hundred twenty-nine stunted children (height-for-age < -2 SD of the National Center for Health Statistics references) aged 9 to 24 months were randomly assigned to one of four groups: control, stimulation, supplementation, or both treatments combined. A fifth group of non-stunted children from the same neighbourhoods was also included. The treatments were delivered weekly for two years. The child development activity programme was similar to that used previously, and the supplement was 1 kg of full-cream powdered milk. The control group was visited every week to control for any benefit of the extra attention given to the supplemented children.

The developmental levels of all the treatment groups improved relative to the controls. The group that received both treatments showed the greatest improvements and was the only one to catch up to the non-stunted children (fig. 3). The benefits of the combined interventions were additive and not interactive. The children were evaluated four years after the intervention ended. A small global benefit remained from supplementation and/or child development activities, since these groups had higher scores than the controls in 13 or 14 of the 15 tests, more than would have been expected by chance, although the differences in the individual tests were not statistically significant. Factor analysis of the tests yielded three factors. The scores on one of them, the perceptual-motor factor, were significantly higher in children who had participated in child development activities [39]. Unfortunately, the benefits from child development activities and supplementation were no longer additive.

FIG. 3. Mean developmental quotient (DQ) scores of five groups of children over two years. The groups are non-stunted children, and stunted children who received both stimulation and supplementation, supplementation alone, stimulation alone, and no intervention (control) [38]

Comments on the home-visiting strategy

The main function of the home visitors was to encourage the mothers to engage in activities that would help their children’s development. Such skills as questioning, conversation, problem solving, creativity, and explanation were encouraged. In order to do this, the visitors provided the parents with information about the intellectual development of children and teaching strategies that would help promote learning. In addition, guidelines were given as to the content of what could be taught.

During the home visits, activities and concepts that were appropriate for the developmental levels of the child were introduced. The home visitor demonstrated these activities in such a way that they could be clearly understood by both mother and child and would enable the mother to carry out the activities on her own once the visitor had left. The relationship between the community health aide and the mothers was critical, and it was important that both mother and child enjoy the visits.

The choice of the visitor was therefore important to the success of the programme. Some studies elsewhere have used professionals, whereas others have opted for volunteers or paraprofessionals or previous programme graduates. Professionals were usually graduates with training in child development. The use of professionals makes the programme less generalizable, since not only are they scarce but few developing countries could afford to pay them. Paraprofessionals are less costly than graduates, and they are now being used more often in home-based programmes. A certain level of literacy is required, since the programmes often involve structured curricula that demand reading as well as record keeping. Some of the less able community health aides had difficulty in grasping the theoretical concepts behind the programme or in appreciating the developmental stage of the child. We found that it was quite possible for a community health aide to miss the point of an activity or to forget to link the appropriate concept with the activity. These problems were overcome only by constant supervision, training, and reinforcement.

The use of paraprofessionals had the advantages that they were readily available and the cost was low. In addition, they were comfortable working in poor neighbourhoods and could readily understand the family’s problems and relate to them easily.

Child-to-child model

In addition to the home-visiting approach, we developed a child-to-child model that was targeted at children from remote rural areas where home visiting was thought to be less feasible. The aim was to work at school with the target children’s older siblings, who would then take the intervention home. Older children traditionally help with the care of their younger siblings, so this was culturally appropriate and feasible if the children attended school. A pilot study was conducted to improve stimulation in the home and introduce some health practices. The schoolchildren made toys for their younger siblings and were taught how to play with them at home. After one year in the programme, the schoolchildren’s knowledge of child development had improved and the mothers reported that they brought the toys home and played with their siblings. However, there was no improvement in the siblings’ development [40, 41]. We concluded that the inputs were too small to affect the younger children’s development. The school-children themselves were the main beneficiaries of the programme, and the Ministry of Education has modified the ideas generated by the programme and included them in the primary school curriculum.

Summary and conclusions

Developed countries
Developing countries

Developed countries

A large number of studies on the effects of early intervention on the development of poor children have been carried out. Some of those with good study designs and evaluations have been reviewed above. They were also chosen to reflect a variety of programme models that have been tried. Although problems such as attrition of the sample can make interpretation difficult, some conclusions can be drawn.

Well-designed early childhood care and education programmes can produce significant benefits to the participants’ development that can influence later school performance and social outcomes that prepare them for adulthood. The benefits varied with the type of intervention. Centre-based programmes were more consistent in showing improvements in the IQs of the children than home-based programmes. There was some suggestion that the more intensive programmes that involved a greater amount of time spent with the children produced larger IQ benefits to the children.

Few of the programmes that began in infancy reported benefits before two years of age. Presumably the scores of the control groups were still high, and the decline in development commonly seen in poor children had not yet started. By three years of age, more consistent differences began to emerge between programme children and controls. However, it is not possible to determine whether the first two years of intervention added to these benefits. Programmes with older children that were designed to investigate the effects of age of entry did not find any significant differences between the benefits to children entering at three years of age and the benefits to those entering at age four.

The results from programmes that included long-term follow-up showed that, in general, it was possible for benefits to IQ to remain up to three or four years after the end of the programme. The Carolina Abecedarian Programme reported persistence in IQ gains up to the age of 12 years. Benefits in other areas that could influence performance in school were also seen. There was less need for special education and less repetition of grades among programme participants.

Many studies also reported benefits to the mother that helped them to play a more supportive role in their children’s development and also benefits to themselves, such as better self-esteem and a sense of self-worth.

Developing countries

The main problem with most of the studies from the United States was that they were expensive, particularly those which were centre based. These models would be neither feasible nor practical in developing countries without substantial modifications to reduce costs. Many innovative programmes in developing countries have been conducted, and benefits have usually been reported. However, evaluations have seldom been rigorous or long term. Tightly controlled research projects in developing countries have resulted in important concurrent benefits to the children’s development, and some benefits have been sustained at least for a few years. However, the studies are few in number and generally too expensive to be copied on a large scale.

The Jamaican series of studies used a slightly less expensive model (although still probably only feasible for small, targeted populations). These studies showed that teaching child development activities to mothers in a home-visiting programme was successful in improving their children’s development in the short term and had some sustained benefits. It was possible to use paraprofessionals, but a minimum frequency of visiting was required. A child-to-child approach working with schoolchildren was less likely to improve the development of younger siblings.

Another concern in developing countries is the poor quality of the primary schools. In many instances, these are likely to erode benefits gained in the early childhood programmes. Therefore, it is probably necessary to continue interventions at least through the first grade of primary school. The challenge is to find low-cost ways of doing this.


1. Gaber H. The Milwaukee Project: preventing mental retardation in children at risk. Washington, DC: American Association on Mental Retardation, 1988.

2. Campbell FA, Ramey CT. Effects of early intervention on intellectual and academic achievement: a follow-up study of children from low-income families. Child Dev 1994;65:684-98.

3. Wasik BH, Ramey CT, Bryant BM, Sparling JJ. A longitudinal study of two early intervention strategies: Project CARE. Child Dev 1990;61:1682-96.

4. Andrews SR, Blumenthal JB, Johnson DL, Khan AJ, Ferguson CJ, Lasater TM, Malone PE, Wallace DB. The skills of mothering: a study of parent child development centers. Monogr Soc Res Child Dev. Chicago, 111, USA: University of Chicago Press, 1982;47(6).

5. Karnes MB, Teska JA, Hodgins AS, Badger ED. Educational intervention at home by mothers of disadvantaged infants. Child Dev 1970;41:925-35.

6. Infant Health and Development Program (IHDP). Enhancing the outcome of low birth weight, premature infants: a multi-site randomized trial. JAMA 1990;263: 3035-42.

7. Lambie DZ, Bond JT, Weikart DP. Home teaching with mothers and infants. The Ypsilanti-Carnegie Infant Education Project: an experiment. Monographs of the High/ Scope Educational Research Foundation No. 2. Ypsilanti, Mich, USA: High/Scope Press, 1974.

8. Jester RE, Guinagh BJ. The Gordon parent education infant and toddler program. In: Consortium for Longitudinal Studies. As the twig is bent. Lasting effects of preschool programs. Hillsdale, NJ, USA: Lawrence Erlbaum, 1983:103-32.

9. Gray SW, Ruttle K. The family-oriented home visiting program: a longitudinal study. Genet Psychol Monogr 1980;102:299-316.

10. Gutelius MF, Kirsch AD, MacDonald S, Brooks MR, McErlean T, Newcomb C. Promising results from a cognitive stimulation program in infancy. Clin Pediatr 1972;11:585-93.

11. Gray SW, Klaus RA. The early training project: a seventh-year report. Child Dev 1970;41:909-24.

12. Gray SW, Ramsey BK, Klaus RA. The early training project 1962-1980. In: Consortium for Longitudinal Studies. As the twig is bent. Lasting effects of preschool programs. Hillsdale, NJ, USA: Lawrence Erlbaum, 1983:33-69.

13. Schweinhart LJ, Weikart DP. Young children grow up: the effects of the Perry Preschool Program on youths through age 15. Monographs of the High/Scope Educational Research Foundation No. 7. Ypsilanti, Mich, USA: High/Scope Press, 1980.

14 Schweinhart LJ, Weikart DP. The effects of the Perry Preschool Program on youths through age 15 - a summary. In: Consortium for Longitudinal Studies. As the twig is bent. Lasting effects of preschool programs. Hillsdale, NJ, USA: Lawrence Erlbaum, 1983:7-101.

15. Levenstein P. Cognitive growth in preschoolers through verbal interaction with mothers. Am J Orthopsychiatr 1970;40:426-32.

16 Levenstein P, O’Hara J, Madden J. The mother-child home program of the verbal interaction project In: Consortium for Longitudinal Studies. As the twig is bent. Lasting effects of preschool programs. Hillsdale, NJ, USA: Lawrence Erlbaum, 1983:237-63.

17. Madden J, Levenstein P, Levenstein S. Longitudinal IQ outcomes of the mother-child home program. Child Dev 1976;47:1015-25.

18 Lazar I, Darlington R. Lasting effects of early education. A report from the consortium for longitudinal studies. Monogr Soc Res Child Dev. Chicago, 111, USA: University of Chicago Press. 1982:47(2-3).

19. Barnett SW. Long-term effects of early childhood programs on cognitive school outcomes. Future Child 1995;5:25-50.

20. Ramey CT, Ramey SL. Early intervention and early experience. Am Psychol 1998;53:109-20.

21. Boocock SS. Early childhood programs in other nations: goals and outcomes. Future Child 1995;5:94-114.

22. McKay H, Sinesterra L, McKay A, Gomez H, Lloreda P. Improving cognitive ability in chronically deprived children. Science 1978:200:270-8.

23. Waber DP, Vuori-Christiansen L, Ortiz N, Clement JR, Christiansen NE, Mora JO, Reed RB, Herrera MG. Nutritional supplementation, maternal education and cognitive development of infants at risk of malnutrition. Am J Clin Nutr 1981;34:797-803.

24. McKay A, McKay H. Primary school progress after preschool experience: troublesome issues in the conduct of follow-up research and findings from a Cali, Colombia study. In: King K, Myers R, eds. Preventing school failure: the relationship between preschool and primary education. Ottawa: International Development Research Center, 1983:32-42.

25 Halpern R, Myers R. Effects of early childhood intervention on primary school progress and performance in the developing countries. Report prepared for the Bureau for Program and Policy Coordination, USAID. Ypsilanti, Mich, USA: High/Scope Foundation, 1985.

26. Myers R. The twelve who survive. London: Routledge 1992.

27. Kagitcibasi C. Parent education and child development In: Young ME, ed. Early child development: investing in our children’s future. Proceedings of a World Bank Conference. Amsterdam: Elsevier, 1997:243-72.

28. Kagitcibasi C, Sunar D, Bekman S. Comprehensive preschool education project: final report. Istanbul: Bogazici University 1987.

29. Chaturvedi E. Impact of six years exposure to ICDS scheme on psychosocial development. Indian J Pediatr 1987;24:153-60.

30. Walker SP, Grantham-McGregor SM. Growth and development of West Indian children. Part 2: Development. West Indian Med J 1990,39:12-19.

31. Wein N. Analysis of four subtests of the Caldwell preschool inventory revised for Jamaican children. Report from the Bernard Van Leer Foundation of Jamaica. Kingston, Jamaica: 1971.

32. Chambers C, Grantham-McGregor SM. Patterns of mental development among young middle-class Jamaican children. J Child Psychol Psychiatry 1986;27:117-23.

33. McDonald K, Grantham-McGregor S, Chang S. Social stimulation of the severely malnourished child: a home training program. Indian J Pediatr 1989;56:97-103.

34. Grantham-McGregor SM, Desai P. A home-visiting intervention program with Jamaican mothers and children. Dev Med Child Neurol 1975;17:605-13.

35. Powell C, Grantham-McGregor SM. Home visiting of varying frequency and child development. Pediatrics 1989;84:157-64.

36. Grantham-McGregor SM, Schofield W, Powell C. Development of severely malnourished children who received psychosocial stimulation: six year follow-up. Pediatrics 1987;79:247-54.

37. Grantham-McGregor SM, Powell C, Walker S, Chang S, Fletcher P. The long-term follow-up of severely malnourished children who participated in an intervention program. Child Dev 1994;65:428-39.

38. Grantham-McGregor SM, Powell CA, Walker SP, Himes JH. Nutritional supplementation, psychosocial stimulation and mental development of stunted children: the Jamaican study. Lancet 1991;338:1-5.

39. Grantham-McGregor SM, Walker SP, Chang SM, Powell CA. Effects of early childhood supplementation with and without stimulation on later development in stunted Jamaican children. Am J Clin Nutr 1997;66:247-53.

40. Knight J, Grantham-McGregor SM, Ismail S, Ashley D. A child-to-child program in rural Jamaica. Child: Care, Health and Development 1991;17:49-58.

41. Knight JL. An experimental study of sibling influence on the development of preschool children. Master’s Degree Thesis, University of the West Indies, Mona, Kingston, Jamaica, 1983:100-1.

Previous Page Top of Page Next Page