HOME scores

Vimpani GV, Baghurst PA, Wigg NR, et al. 1989. The Port Pirie cohort study—cumulative lead exposure and neurodevelopmental status at age 2 years Do HOME scores and maternal IQ reduce apparent effects of lead on Bayley Mental scores In Smith M. Grant LD, Sors A, eds. Lead exposure and child development An international assessment. Lancaster, UK Kluwer Academic Press. 

The BAEP measures which produced the most significant results in the 5-year follow-up were found to be associated with HOME scores (which were controlled in the data analysis), and sex differences were noted in BAEP latencies. 

It is clear from the work of Otto and colleagues, where BAEP measures were found to be related to HOME scores, that the culture-free nature of EEG measures cannot be assumed. It is likely that there are a number of influences on EEG measures. It is known, for example, that diet can affect EEG measurements. 

The socioeconomic status (SES) of the infant s family was assessed at three months postpartum with the Hollingshead Four-Factor Index of Social Status (Hollingshead, 1975). The mean SES for families in this sample was 17.1 (SD = 5.4), reflecting the preponderance of single-parent low-income households. At 6 and 12 months, the quality of the infant s domestic environment was assessed with Caldwell and Bradley s Home Observation for Measurement of the Environment (HOME) (Caldwell and Bradley, 1978). The HOME combines interviewers questions with direct observations to yield subscale and total HOME scores. In general, the HOME measures both the parent s responsivity and personal involvement with her infant and the extent to which the physical and temporal environment is stimulating and safe. The mean of the 6- and 12-month HOME scores was calculated for each subject and tested as a potential developmental covariate or confounder in regression analyses. 

The Port Pirie Cohort Study - Cumulative Lead Exposure and Neurodevelopmental Status at Age 2 Years Do HOME Scores and Maternal IQ Reduce Apparent Effects of Lead on Bayley Mental Scores ... 

The total HOME score modal value was 44, which is somewhat higher (more favourable) than the mean score in the standardizing sample of Bradley and Caldwell (37.54 with a SD of 10.4). The distribution was significantly skewed to the left. 

There was a strong positive correlation between the HOME score and the Bayley MDI. These two measures were negatively correlated with postnatal PbB concentrations (expressed as integrated PbB which is derived by trapezoidal integration of each individual s blood lead curve derived from PbB estimates of cord, 6-, 15- and 24-month samples) (Table 6). A very low HOME score also appeared to be associated with a low Bayley PDI, but no significant relationship was found between these two measures across the rest of the HOME score range. 

Table 6 Mean Bayley MDI at 2 years and geometric mean of integrated PbB according to total HOME score at 3 years...

The association of a number of maternal, sociodemographic and perinatal variables with total HOME score was evaluated before assessing the effect of including the child s HOME score at age 3 years along with these variables in a multivariate analysis examining the explanatory power of the various PbB measures. 

A clearly positive association between mother s age, years of secondary education of both parents, mother s and father s workplace, parental relationship, marital status and maternal WAIS and the total HOME score was observed. Perinatal variables - with the exception of oxygen use which was of borderline significance - had no statistically significant relationship to HOME score (Table 7). 

A second multiple regression analysis was undertaken in which PbB measures were entered as explanatory variables before and after HOME scores and maternal WAIS had been entered in the model. This analysis strategy is conservative in that it assumes that the effects of all the non-lead variables are associated with MDI via independent (i.e. direct) mechanisms that do not involve lead exposure (i.e. they are pure confounders as shown in Model 1, Figure 3). In reality, it is plausible - even likely - that part of the effect of some non-lead variables is mediated via altered exposure to lead. 

There are two other possible explanations for the observed relationships between PbB, HOME scores, other explanatory sociodemographic and perinatal variables, and Bayley MDI scores (Figure 3). 

The possibility of this explanation was investigated by examining the relationship between HOME scores and Bayley MDI in children with high, intermediate and low integrated PbB levels to see whether the association between HOME and MDI was consistent across the whole spectrum of lead... 

Table 7 Mean HOME score tabulated by sociodemographic factors, maternal IQ and perinatal factors...

Table 8 Partial linear regression coefficients of MDI with each blood lead measure and without maternal IQ and HOME score in the modeP...

A second explanation is that the HOME score is the main predictor (i.e. effector variable) of Bayley MDI, and the observed association of PbB with MDI is found because of the high correlation between PbB and HOME (Model 3, Figure 3). In these circumstances, PbB could be considered as a surrogate for HOME. If, indeed, total HOME score is highly correlated with PbB it would be difficult to demonstrate much additional effect of PbB on Bayley MDI in the regression analysis (after allowing for HOME score), even if it did have some independent effect. It is clear from preliminary analyses that these variables are correlated to an extent which makes such considerations important. 

Table 9 Relationship between HOME score and Bayley MDI according to integrated PbB n in brackets)...

HOME score Bottom qmrtile (PbB 23pgdl- )... 

---