Sirs,
Birthweight is not only a marker of current inequalities but, potentially, an
important indicator of progress in reducing inequalities in health, irrespective
of the pathways that lead to lower birthweight. In addition to numbers of low-birthweight
babies, we suggest that birthweight gradient in relation to proxies for poverty
might be a useful way of monitoring the impact of poverty and the progress of
interventions, both medical and political, on the vitally important indicator of
birthweight.
We have previously shown that there are significant errors in allocating postcodes to enumeration districts.' We agree with Aszkenasy and Hutchison2 that data supplied by local authorities offer more precise allocation to postcode and are updated each year, thus avoiding the errors caused by changing demographics between Censuses. In Sheffield we obtained data for income support and housing benefit in terms of the total number of households and the proportion of households receiving these benefits. 3 This produced an estimate of deprivation, or poverty index, attached to each postcode.
Data on birthweight matched to postcode were collected from December 9 991 to November 1 996 inclusive. For the purposes of this study we used only the `poverty index' data supplied to us in 1996, as this was the first year in which such data were readily available. The relationship between birthweight and income support was investigated by linear least-squares regression analysis using the GLIM4 statistical computer package. A nominal level of 5 per cent statistical significance was assumed. The adequacy of the regression models was determined by residual examination. For each model, the residuals followed an appropriate normal distribution.
The mean birthweight for each year is presented in Table l; there was little variation over time. The relationship between income support, birthweight and year is presented in Table 2. All the regression coefficients (represented by the slope of the lines) were negative; i.e. the greater the proportion of families on income support, the lower the birthweight. The gradient was steepest for the 1991 cohort. The least steep gradient was for the 1996 cohort. It appears that the gradient is becoming less steep over time (Table 2), Figure 1 shows the same information graphically.
The data supplied rely only on income support and housing benefit, and do not attempt to measure other parameters of deprivation, but have the merit of simplicity and stability over time. Collection and manipulation of such data are part of everyday practice and do not need substantial extra resources.
Many reports on poverty and birthweight use average birthweight or proportions of low-birthweight babies as their measure.' The gradient across income groups offers an alternative approach. Flattening of this gradient over time might suggest a reduction in inequalities, and a steepening of the gradient would suggest the reverse. Similar analyses might be undertaken using height at age five, and obesity measures at a range of ages.
Table 1 Average birthweight in Sheffield (1991-1996)|
Birthweight
(g) |
Table 2 Association between birthweight and income support
Year Slope 95% CI 1991 -3.34 -6.48, -0.20 1992 -3.3 -4.12, -2.46 1993 -2.83 -3.67, -1.98 1994 -2.77 -3.62, -1.82 1995 -2.37 -3.22, -1.52 1996 -1.69 -2.55, -0.83 |
Figure 1 Birthweight and income support from 1991 to1996: changing gradient over time.
|
1 Collins SE, Haining RP, Bowns IR, et al. Errors in postcode to enumeration district mapping and their effect on small area analyses of health data. J Publ Hlth Med 1998; 20(3): 325-330.
2 Aszkenasy M, Hutchison 5. Births, gestation and birthweights in South Tees, 19911-1996. J Publ Hlth Med 2000; 22: 457-451.
3 Crofts DJ, Bowns M, Williams TS, el al. Hitting the target--the equitable distribution of health visitors across caseloads. J Publ Hlth Med 2000; 22:295-301.
4 Spencer NJ, Logan S, Gill L. Trends and social patterning of birthweight in Sheffield, 1985-94. Arch Dis Child 1999; 81: F138-F140,
Yours faithfully,
David M. S. Hall, Professor of Community Paediatrics
Alan S. Rigby, Senior Lecturer in Statistics and Epidemiology
Ian R. Howns, Senior Research Fellow
T. Sim Williams, Research Associate
Debbie Crofts, Research Health Visitor
University of Sheffield SCHARR
Community Sciences Centre,
Northern General Hospital,
Sheffield S5 7AU
|
If you have come to this page from an outside location click here to get back to mindfully.org |