Table 1 Comparison of snapshot and constant-age tracking trends in IDEA autism prevalence for 7 to 10-year intervals beginning in birth years 1988 (interval 1) and 1993 (interval 2)

Interval 1 (snapshot based on 2005 report) Interval 2 (snapshot based on 2010 report)
b snap b track $\frac{{\mathbit{b}}_{\mathbit{snap}}}{{\mathbit{b}}_{\mathbit{track}}}\mathbf{%}$ Birth N b snap b track $\frac{{\mathbit{b}}_{\mathbit{snap}}}{{\mathbit{b}}_{\mathbit{track}}}\mathbf{%}$ Birth N Track Min.
10 -4 y -1 10 -4 y -1   Year   10 -4 y -1 10 -4 y -1   Year   Age Snap
Span   Span   Age
4.1 ± 1.5 5.3 ± 2.1 78 ± 8 1988-1997 24 6.3 ± 2.1 8.6 ± 2.6 73 ± 8 1993-2002 19 8 8
4.4 5.4 82   1 6.8 8.7 78   1
4.2 ± 1.6 5.6 ± 2.2 77 ± 8 1988-1996 22 6.2 ± 2.2 8.5 ± 2.7 73 ± 8 1993-2001 22 9 9
4.5 5.5 81   1 6.9 8.7 80   1
4.6 ± 1.6 6.0 ± 2.2 78 ± 7 1988-1995 21 6.6 ± 2.3 9.0 ± 2.9 74 ± 8 1993-2000 18 10 10
4.7 5.7 81   1 7.0 8.8 80   1
5.2 ± 1.8 6.7 ± 2.4 77 ± 8 1988-1994 17 7.5 ± 2.9 9.8 ± 3.4 76 ± 8 1993-1999 11 11 11
4.8 6.0 81   1 7.0 8.6 81   1
1. Snapshot slopes for interval 1 and 2 are based on the 2005 and 2010 IDEA reports, respectively. Slopes (symbol b, in units of # 10-4 y-1) are estimated based on least squares linear regression. The bottom line (N = 1) in each age pair shows results for California. The top line shows the mean and standard deviation for all states (N = 11 to 24) for which the error in the constant-age tracking slope is < 10% and the error in the snapshot slope is < 15%. The snapshot:tracking slope ratio is reported as a quantitative estimate of the fraction (in%) of the constant-age tracking slope that is "real" rather than due to better and expanding diagnosis. A range of tracking ages from 8 to 11 is considered.