Small area variation in low birthweight: looking beyond socioeconomic predictors.

The main objectives of this study were to determine the geographic distribution of low birthweight rates in London. Ontario and to identify small areas within the city that have low birthweight rates not adequately explained by the areas' socioeconomic characteristics. The following socioeconomic variables were used in a weighted, ecological, multiple regression analysis; % unwed mothers, % teen mothers, % low income, % low education, % unemployment and % immigrants. The overall variation in low birthweight rates was statistically significant and largely, but not entirely, explained by the socioeconomic characteristics of the areas. Two out of 31 census tract clusters were identified as having low birthweight rates which were higher than would be expected based on their socioeconomic profile. This methodologic approach may interest health planners as it draws attention to local factors other than socioeconomic ones which may be important when developing local strategies for low birthweight prevention.

Randomized trial of the effect of calcium supplementation on fecal risk factors for colorectal cancer.

Newmark et al. (J Natl Cancer Inst 1984; 72:1323-8) hypothesized that supplementation with calcium would decrease the concentration of bile acids in aqueous phase feces and that such a reduction would reduce the risk of malignant disease in the bowel. A randomized trial was therefore undertaken to examine the effects of calcium supplementation on fecal biochemistry. A total of 68 men between 40 and 60 years of age volunteered to participate after having been selected randomly from population lists in Scarborough, a city in the Toronto area, Ontario, Canada. Participants were randomly assigned to receive a daily supplement of 3 g of calcium carbonate or a sucrose placebo for a period of 1 week. Fecal samples were collected for 2 days prior to supplementation and for the last 2 days of supplementation. Records of all foods consumed were kept throughout the study period. The average concentration of total soluble bile acids fell in the placebo group (-11.2 g/ml) but increased slightly in the calcium group (1.4 g/ml). Similar patterns of change were observed for deoxycholic acid (placebo, -3.0 g/ml; calcium, 4.5 g/ml). The distribution of changes in total bile acids and deoxycholic acid differed between randomization groups at the 10 percent level of significance in univariate analysis. After adjustment for initial stool chemistry and initial levels and changes in nutrient intake, no reduction in fecal bile acid levels was observed in association with calcium supplementation. In fact, a statistically significant (p = 0.05) increase in deoxycholic acid concentration remained in the calcium supplemental group. Thus, this study was unable to support the hypothesis that calcium supplementation alters aqueous phase bile acids in the manner hypothesized to be consistent with protection from colorectal cancer.

An analysis of blood lead data in clinical records by external data on lead pipes and age of household.

This study examined the possibility that lead pipes in the drinking water distribution system were elevating the blood lead levels of children in London, Ontario, Canada. Based on their postal codes, 164 children admitted between 1984 and 1989 to an institution for the behaviorally disordered or developmentally challenged were categorized according to whether they lived in the area of the city known by the local Public Utilities Commission to be serviced by lead pipes. Analysis of covariance was used to obtain confounder-adjusted geometric means in each area. After adjusting for gender, year of lead test (a surrogate for gasoline source), and census tract prevalence of low family income, children in the lead service area (LSA) were found not to have higher blood lead levels (geometric means: LSA = 4.7 micrograms/dl, Non-LSA = 4.8 micrograms/dL; p = 0.839). The average blood lead level declined 60.9% between 1984 and 1989. Using municipal tax assessment data on the age of each child's home, those children living in homes built during or before 1945 (when interior paints were as much as 50% lead by dry weight) had an average blood lead level that was 62.3% higher (p = 0.011) than that of those in homes built since 1975 (when interior paints were limited to no higher than 0.5% lead by dry weight). A clear gradient was observed. This association with age of home remained significant after adjusting for gender, diagnosis, and year of lead test. Variables indicating the amount of industry near the child's residence and the presence of lead service pipes did not enter the model after house-age. In conclusion, no evidence indicated that the lead service pipes were elevating blood lead levels in these London children. The data suggest that with the removal of lead from gasoline, lead-based paint is a significant remaining source of lead exposure. Little data are available on childhood lead exposure from paint in Canada. The present descriptive data suggest that more research into this potential problem in Canada is warranted.