Table 3.
Effect of nondifferential exposure misclassification bias on the estimated ORs for the association between selected months of conception and autism before 6 years of age.
| Month of conception | Observed results | Underlying True Resultsa | Difference between observed OR and true OR (bias) | ||||
|---|---|---|---|---|---|---|---|
| Autism | OR (95% CI) | Autism | OR (95% CI) | ||||
| Yes No. | No No. | Yes No. | No No. | ||||
| January | 1775 | 581,406 | 1.1337 (1.05 – 1.21) | 2367 | 580,814 | 1.1340 (1.06 – 1.20) | −0.0003 |
| February | 1507 | 470,667 | 1.1890 (1.10 – 1.27) | 2009 | 470,165 | 1.1890 (1.11 – 1.26) | −0.0000 |
| March | 1944 | 593,302 | 1.2167 (1.13 – 1.30) | 2592 | 592,654 | 1.2170 (1.14 – 1.29) | −0.0003 |
| Julyb | 1473 | 547,009 | 1.000 | 1964 | 546,518 | 1.000 | |
a
Assuming 25% of cases are misclassified independently of month of conception.
b
Reference month.