Error corrected, conclusions the same
CMAJ 1997;157:646
See also:
In the article "Recent trends in infant mortality rates and proportions of low-birth-weight live births in Canada" (CMAJ 1997;157:535-41 [abstract / résumé]), Drs. K.S. Joseph and Michael S. Kramer identify a possible error in the birth weight data from Ontario for 1993 and 1994. The error was traced to improper keying (data capture) of a small number of records. For birth weights reported in pounds and ounces, the second digit of the ounces was omitted, e.g., 5 pounds 10 ounces became 5 pounds 1 ounce.
Since we were made aware of this problem, the data have been rekeyed to correct the error for 1993 and 1994. As well, 1995 data are now available. Finally, to validate the vital statistics data, they were compared with birth weight data from hospital discharge abstracts, which are available up to 1994. The same statistical tests used by Joseph and Kramer were conducted on the revised data to determine whether there was a significant increase in the proportion of low-birth-weight babies born in Ontario between 1987 and 1995.
The corrections to the data reduced the percentage of low-birth-weight babies from 6.06% to 5.87% in 1993 and from 6.54% to 5.93% in 1994 (Table 1). The 1995 percentage is 5.98%. The 199495 to 1987 ratio is, therefore, 1.11. But, despite the reduction in the 1993 and 1994 percentages, the statistical tests show that the increase in the proportion of low-birth-weight babies from 1987 to 1995 is still statistically significant. The hospital discharge data also indicate a statistically significant trend.
The differences in the proportion of low-birth-weight babies between the vital statistics and the hospital discharge data can be explained by 2 factors: (1) vital statistics include all births among Ontario residents, whereas hospital discharge data include only hospital births and only births among Ontario residents occurring within Ontario; (2) in the vital statistics, birth weights are reported by the mothers, whereas those in hospital discharge data are reported by the attending physicians.
Joseph and Kramer showed a significant increase in low-birth-weight babies in Ontario, but the results were somewhat constrained by the incorrect data. The corrected data, however, still indicate a significant increase. The trend is confirmed by the hospital discharge data. At the national level, the percentage of newborns of low birth weight was 5.78% in 1995, up from 5.44% in 1992.
Indira Singh
Deputy Registrar General
Ontario Ministry of Consumer and Commercial Relations
Toronto, Ont.
Janet Hagey
Director
Health Statistics Division
Statistics Canada
Ottawa, Ont.
 |
Send a letter to the editor responding to this letter
Envoyez une lettre à la rédaction au sujet de cette lettre |
Letters