TABLE 2.

Linear Regression Predicting Age of Diagnosis in Years

Coefficient95% Confidence Interval
Diagnosis
    Asperger's disorder1.80.8 to 2.9
    PDD0.2−0.1 to 0.6
    Interaction of Asperger's diagnosis and age0.40.3 to 0.5
Demographics
    Male0.2−0.2 to 0.6
    Age, y0.20.1 to 0.2
    Black0.1−0.7 to 0.4
    Asian/Pacific Islander−0.5−1.5 to 0.4
    Native American0.0−0.9 to 0.9
    Latino−0.8−1.6 to 0.1
    Child was adopted0.80.2 to 1.5
    Rural0.40.1 to 0.7
    Suburban0.2−0.2 to 0.6
    Income below poverty level0.2−0.2 to 0.7
    Income from poverty level to 100% above0.90.5 to 1.2
Symptoms
    Does not respond when called0.0−0.4 to 0.4
    Self-injurious−0.2−0.5 to 0.1
    Severe language deficits1.2−1.5 to −0.8
    Oversensitive to pain0.60.2 to 0.9
    Aggressive to others−0.1−0.4 to 0.2
    Insists on sameness−0.1−0.4 to 0.3
    Sustained odd play0.3−0.6 to −0.1
    Echolalia−0.2−0.5 to 0.0
    Hand flapping0.4−0.7 to −0.1
    Toe walking0.2−0.5 to −0.1
    Spins self−0.1−0.4 to 0.2
Other clinical features
    Mental retardation−0.2−0.6 to 0.2
    Hearing impairment0.80.1 to 1.6
    Seizures0.0−0.5 to 0.5
Health system characteristics
    2 primary care physicians before diagnosis0.0−0.3 to 0.3
    3 primary care physicians before diagnosis0.0−0.4 to 0.4
    ≥4 primary care physicians before diagnosis0.50.1 to 1.0
    In response to concerns about development
        Physician conducted developmental tests−0.2−1.0 to 0.6
        Physician referred to a specialist0.3−0.7 to 0.0
    Interaction of developmental testing and referral0.3−0.6 to 1.3
  • Adjusted R2 = 0.54. Statistically significant findings at P ≤ .05 are in bold. Reference group consists of female white children who received a diagnosis of autistic disorder, and live in urban settings with household incomes >100% above the federal poverty level.