engleski

Parental education as a confounder in the assessment of low level lead effect on psychological functions in children

A cross-sectional study was performed on 275 pupils (144 boys and 131 girls) from the third and fourth grade of three elementary schools in Zagreb, Croatia, in order to provide relevant data on the relationships between characteristic indicators of lead exposure / health effects including hematological condition and some psychological functions. According to school locations, traffic density was the lowest in the street near school 1 and highest in the vicinity of school 2, while traffic density near school 3 ranged in density between school 1 and school 2. Air lead concentration was measured continuously in the vicinity of school 2, where the average value never exceeded 1 micro g Pb/m^3. In each pupil the intensity of lead absorption was determined by analyses of blood lead (PbB), erythrocyte delta-aminolevulinic acid dehydratase (ALAD), and erythrocyte protoporphyrin (EP). Lead in erythrocytes (PbE) was calculated by the PbB and corresponding hematocrit (Hct) value. The hematological condition of pupils was tested by hemoglobin (Hb) and Hct determination. Psychological examination included assessment of general cognitive development, visual motor integration, attention and complex psychomotor reaction. General intellectual status was evaluated by Wechsler Intelligence Scale for Children - Revised. Verbal subtests were: Information, Comprehension, Arithmetic, Similarities and Digit span, which were used to estimate verbal intelligence quotient (VIQ). Nonverbal subtests were: Picture completion, Picture arrangement, Block design, Object assembly and Coding, which were used to estimate nonverbal intelligence quotient (NIQ). All ten subtests were used to estimate general intelligence quotient (IQ). Visual motor integration was estimated by Bender Gestalt Test (BG) with an objective Göttingen scoring system. The quality of performance was scored by the number of errors in nine drawings. Attention and complex psychomotor reactions were assessed using the Complex Reactionmeter Drenovac. The total time (in ms) was determined by measuring simple psychomotor reaction to sound stimulus (S) and to light stimulus (L), as well as complex reaction time (4A) by measuring different complex psychomotor reaction. In each test the number of errors (SE, LE, 4AE) was also determined. Both parent and teacher ratings of child behavior were assessed by means of an extensive questionnaire. The socio-economic background of the family was assessed through the parent's education status, i.e. exact length of education for each parent, that may act as proxy for parents' IQ. Means and their 95% confidence limits were compared among the schools. Multiple stepwise regression analysis was performed for each of the 20 measured dependent psychological variables, standardized for school and gender, with age and six indicators of lead absorption and hematological condition as independent variables. Criterion for entry was p<0.05, and for retention p<0.1. The summarized results (N=275) for indicators of lead absorption (Mean and 95% CI - PbB: 70.8, 68.6 to 72.9 micro g/L ; PbE: 179.2, 173.9 to 184.6 micro g/L ; ALAD: 59.8, 58.6 to 61.5 U/L E ; EP: 1.26, 1.22 to 1.30 micro mol/L E) showed that lead exposure of pupils can be characterized as relatively low. The anticipated difference in lead exposure with regard to the difference in traffic density in the vicinity of the schools, was verified by the difference in ALAD activity (school 1 > school 3 >school 2) and partly in PbB and PbE value (school 1 < school 2). The values of hematological indicators (Mean and 95% CI - Hb: 136.7, 135.8 to 137.7 g/L ; Hct: 39.5, 39.3 to 39.7) were in the expected range for Croatian children. The summarized results (N=275) of psychological examinations (Mean and 95% CI - VIQ: 112.9, 111.6 to 114.3 ; NIQ: 111.9, 110.4 to 113.4 ; IQ: 113.7, 112.3 to 115.0 ; BG: 7.9, 7.5 to 8.4 ; S: 449.5, 441.5 to 457.5 ms ; L: 578.4, 568.3 to 588.5 ms ; 4A: 740.4, 704.9 to 776.0 ms ; SE: 1.36, 1.15 to 1.56 ; LE: 1.51, 1.30 to 1.73 ; 4AE: 13.67, 12.62 to 14.71) were also in the expected range. In relation to schools, there were no consistent differences in the scores for psychological performance, contrary to what was expected, having in mind the differences in lead absorption. The sequence of scores for VIQ, NIQ and IQ (school 1 < school 3 < school 2) and the sequence of scores for BG (school 1 > school 3 > school 2) were in the opposite direction then expected in relation to the respective lead absorption. These findings indicate that factors other than lead were responsible for differences in psychological performance. When parents' years of education were included with age, lead absorption indicators, and hematological indicators as independent variables in regression analysis with 20 dependent psychological variables, age and parents' education were the only independent variables significantly associated with psychological variables. One or both parents' years of education were significant (p<0.05) predictor for 16 of 20 variables, age was significant for 9, and Hb for 4 of 20 variables. After Bonferroni correction of significance levels, to adjust for multiple regressions, age was significant for 6 variables and parents' education for 9 variables. The nonstandardized regression coefficients of PbB for VIQ, NIQ and IQ were -0.016, -0.031, and -0.025, respectively, all nonsignificant. The results obtained do not support the hypothesis that concurrent Pb levels in the range of 70.8&plusmn ; 17.88 micro g/L impair the intellectual development of 9 to 10 year old children. On the other hand, age and parents' education were the only independent variables significantly associated with psychological variables. The parental education is a very important confounding factor in assessing the harmful effects of low level lead exposure on the intellectual ability of children. Consistent insignificance of association between psychological variables and lead exposure suggests that lead's no effect threshold value, if it exists, must be in the upper range of our data, or above. In this way, our results provide the important information to the knowledge of the no observed effect level (NOEL) for lead.

lead; children; psychological functions

Rad je kao poster prezentiran i na skupu 10th Conference of European Association for Research on Adolescence, održanom od 02.-06.05.2006., Alitalia, Turska ; objavljen u Knjizi sazetaka ; Antalya : Center for Research on Child Culture, 2006. ; str. 25-25.