Table 2:

Estimated coefficients for linear regression model for papillary thyroid cancer incidence

EstimateStd. ErrorT ValuePr (< |t|)
Baseline intercepta178.799.0519.75<.0001
Baseline slope CTb15.60.3939.84<.0001
Diff. intercept sizec
    10–14 mm−52.9312.8−4.13.0001
    15–19 mm−37.712.8−2.94.0043
    20–29 mm29.4112.82.3.0245
    30–39 mm−74.1412.8−5.79<.0001
    >40 mm−77.8712.8−6.08<.0001
Diff. slope CT:sized
    10–14 mm−12.470.55−22.52<.0001
    15–19 mm−8.720.55−15.75<.0001
    20–29 mm−11.50.55−20.77<.0001
    30–39 mm−10.280.55−18.57<.0001
    >40 mm−13.120.55−23.69<.0001
  • Note:—The estimated coefficients indicate how changes in CT imaging volume, tumor size, and interaction between these 2 factors would affect thyroid cancer incidence compared with the baseline number of cases of size < 10 mm in 1993.

  • a Baseline intercept term is the estimated incidence in the year 1993.

  • b Baseline slope CT refers to the rate of increase in cases per million CT scans for tumors < 10 mm since 1993.

  • c Diff. intercept size indicates how much the intercept changes for other sized tumors.

  • d Diff. slope CT:size indicates how much the slope changes for other sized tumors. A negative estimate value indicates that the intercept or slope would be lower than for the baseline case of size < 10 mm.