ECDC's logo
The Belfast field studies

One of the largest cities in the United Kingdom played an active part in promoting the Child Development Programme. Some tens of thousands of Belfast families were involved in this programme for over a decade in which the CDP was the main form of parent support for a significant part of that city’s families, during the 1980s and 1990s.

The Eastern Health and Social Services Board (EHSSB), responsible for providing both health and social services throughout Belfast and its satellite towns, collaborated closely with the ECDC during those years. A particular form of collaboration took place through the use of an evaluation instrument known as the Early Health and Development Monitor, piloted and used by the ECDC as its main form of field evaluation across a number of different health authorities. For this monitor, health visitors gathered information each year, starting from birth, on pre-school children and their home environments. (The ECDC’s creation and development of the monitor is described in one of the other articles appearing on this web page.)

Effects evaluated on large samples    Many tens of thousands of Belfast children, both programme and non-programme, and their parent and home environments, were assessed with the use of this Monitor. It took about 15 minutes to complete in each family once a year, and was filled in by the family’s health visitor. The Board’s IT staff were responsible for inputting the very considerable volume of data from thousands of families.

The Monitor proved invaluable for the Belfast community health management teams. For some years its outcomes were used as bench-marks for breast-feeding and immunisation rates across the four health authorities within the Board. Because the Monitor was, where feasible, used annually in families between birth and the age of five, it became possible to follow trends within different indicators, showing for example (for the child population as a whole) an overall rise in breast-feeding rates annually, but also an overall increase in hospital admissions in the second year of life of first-born children across the city.

Among the studies undertaken with the aid of the Monitor, particular attention was given to evidence on the health, dietary, developmental and other child and family outcomes, comparing families in programme areas with children in control areas. As was the practice with the programme across the participating UK health authorities, the more disadvantaged areas within a health authority were selected for programme visiting, while more advantaged (or less disadvantaged) areas were used as control areas, with the Monitor being used on most families in both programme and control areas. (Not all programme authorities were willing to make use of the Monitor evaluation strategy because of pressure on staff resources.)

Programme children scored better on almost every major outcome, despite initial differences.    A summary of the quantitative results of a number of different Belfast studies showed, as expected, that programme families were generally more disadvantaged, based on a variety of socio-educational and other indicators, with birth statistics showing a higher proportion of birth problems in programme families, compared with controls. Despite these differences, on almost every major outcome the programme children subsequently scored better than the controls. It can be noted that a great deal of the later outcome information on both programme and control families was gathered by health visitors who were not at that stage involved in the programme, and thus their scoring could be seen as unbiased.

A selection of the charts produced from the Monitor data are reproduced below; they point to some of the differences between programme and control samples. The samples chosen for this study were mainly families with young mothers having their first babies.

Belfast delivery data

Initial differences between programme and control families.     The diagram above shows the considerable differences in the birth experiences of mothers who were in the programme samples, compared with mothers in the control samples. For this study there was a total sample of 2,242 ‘primip’ (first pregnancy) mothers in the age range 14 to 21 years. The evidence on the higher level of birth difficulties in programme families is supported by other evidence that the programme samples were on average more disadvantaged than the control samples. This evidence showed inter alia that programme families had lower levels of telephone and car ownership and poorer accommodation, while more programme mothers left school before 17 years of age. Another statistic showed a 60% higher level of stay in intensive care for programme infants after birth, compared with control infants. The purpose of these and other comparisons was to show that the programme sample as a whole was more disadvantaged and in poorer health than the control sample.

In the light of that background, the following diagrams provide evidence on the positive effects of Child Development Programme home visiting on the 1274 programme families recruited for this sample, compared with the outcomes on 968 non-programme families who received only conventional home visiting.

Belfast delivery data

Immunisation differences.     Because of population movement and inability to monitor families at each of the three periods - birth, six months and two years - the sample of families with full immunisation data at all three ages was reduced to 868 children by the age of two (536 programme and 332 controls). Immunisation figures (see diagram above) showed that at each of those three critical periods, programme children were 2 to 3% more likely to have been immunised despite the fact that the programme sample was in many ways more disadvantaged and suffering poorer health than the controls.

Belfast delivery data

Hospitalisation of children in the first three years     The diagram above shows that for the first year of life there was little difference between programme and control children’s hospitalisation, measured as mean days in hospital for hospitalised children. However for the second and third years of life it was noticeable that programme children who were hospitalised spent fewer days in hospital than their control counterparts. For the reasons given above, these sample sizes were declining with each year. However the total number of children involved at all the 1, 2 and 3 year assessments was 416. The graph on the left refers to the total sample including all socio-educational (SES) groups, whereas the graph on the right refers only to the lowest socio-educational group.

Belfast delivery data

Partner’s involvement in support for mother.     The diagram above shows a modest but consistent difference favouring the involvement of programme fathers (in their support for the mothers), compared with control fathers, measured at both 12 and 24 months of the children’s ages. The comparison shows relatively more partner involvement across all socio-educational groups in programme families, compared with control families.

Belfast delivery data

Dietary differences.     Some of the most meaningful differences between samples were identified in the changing quality of the children’s diets. A detailed assessment of seven different dietary constituents was made at 12, 24 and 36 months, following the same samples of children. These dietary constituents included the estimated adequacy of the intakes of energy, animal protein, non-animal protein, wholefood, vegetables, fruit and milk - based on the mother’s recall of each child’s diet over the previous 24 hours.

A study across the combined sample of 1,345 children, with mothers of all ages, showed a clear and consistent difference in the adequacy of each of those seven dietary constituents between the high, medium and low socio-educational sample groups, as measured at the 12 and 36 month dietary assessments. These dietary differences across the socio-educational levels could be expected, but they also proved to be measurable in the analyses, with the medium SES group having adequacy levels between the high and low SES groups. The purpose of this initial comparison was to establish the validity of using the adequacy measures of the seven food constituents as an indicator of dietary quality. (‘Adequacy’ was scored by the health visitor as “too much”, “adequate”, “insufficient” or “none”, with statistical weighting to give the seven constituent scores for each child at each assessment.)

The diagram above shows that for the sample of 450 children of mothers aged 14 to 21 at the time of birth (278 programme and 172 non-programme), for each of the dietary constituents measured at 12 and 36 month assessments, programme children’s mean diets showed a strong upward trend in quality, compared with the control children whose mean adequacy scores either went downwards or rose at a reduced rate compared with programme children.

Other comparisons showed that the programme children remained behind the less disadvantaged controls in language, social and cognitive development at the age of 24 months, but the reported reading to children, the frequency of parental reading and the interest taken by children in books - all measured at 36 months - showed that programme children were slightly ahead of controls, despite the fact that the home (adult) reading environment still favoured the controls.

Summary.     Looking at these findings as a whole, there were many positive effects of the programme on meaningfully large numbers of sample children in the Belfast studies, with strong positive achievements on immunisation levels, reduced days in hospital for the growing children, partner involvement (assisting the mother) and the quality of diets in particular. The programme children did not however catch up with or achieve higher mean scores on language, social or cognitive development at 24 months, although by 36 months (by which time the programme visits had long ceased), the levels of reading to children and their interest in books slightly favoured the programme children. The importance of these findings is that they were achieved by programme samples which were shown at the outset of the study (at the time of the children’s births) to be noticeably more disadvantaged than the control samples.

Ref:  Barker W, Anderson R, Chalmers C, O’Sullivan G, and McAteer D (1994)   EHSSB: health trends over time and major outcomes of the Child Development Programme (82 pps)  ). ECDC in collaboration with the Eastern Health and Social Services Board, Northern Ireland.   This report can be ordered from the ECDC.   See the website’s Contact page. 

Return to page on Research Findings

                                                  To print article

                                                     Hold Ctrl

                                                        Click P