Table 3:

Sex-specific model coefficients for rates of incidentally detected differentiated thyroid cancer by imaging modality, 1998–2017*

CovariateFemale patients; rate ratio (95% CI)
n = 2680
Male patients; rate ratio (95% CI)
n = 1314
Imaging modality
 CTReferentReferent
 Carotid ultrasonography0.15 (0.10–0.23)0.19 (0.12–0.32)
 MRI0.08 (0.05–0.14)0.02 (0.01–0.09)
 SPECT0.16 (0.11–0.24)0.15 (0.08–0.26)
Age group, yr
 15–190.09 (0.06–0.14)0.06 (0.03–0.12)
 20–240.17 (0.13–0.23)0.13 (0.08–0.21)
 25–290.31 (0.25–0.39)0.26 (0.19–0.37)
 30–340.50 (0.42–0.61)0.38 (0.28–0.52)
 35–390.64 (0.54–0.76)0.55 (0.43–0.71)
 40–440.77 (0.65–0.90)0.69 (0.54–0.87)
 45–490.89 (0.76–1.03)0.83 (0.66–1.04)
 50–54ReferentReferent
 55–591.01 (0.86–1.18)1.06 (0.85–1.32)
 60–641.13 (0.96–1.32)1.23 (0.99–1.54)
 65–691.14 (0.97–1.35)1.60 (1.28–1.99)
 70–740.97 (0.81–1.17)1.46 (1.14–1.85)
 75–791.00 (0.82–1.22)1.23 (0.93–1.62)
 80–840.82 (0.65–1.03)1.04 (0.74–1.47)
Diagnosis period
 1998–2002ReferentReferent
 2003–20071.56 (1.30–1.88)1.53 (1.18–1.97)
 2008–20122.61 (2.20–3.09)2.59 (2.05–3.26)
 2013–20173.20 (2.71–3.77)3.31 (2.64–4.14)
Modality (period)
 Carotid ultrasonography (2003–2007)§1.26 (0.77–2.07)0.62 (0.31–1.23)
 Carotid ultrasonography (2008–2012)§1.07 (0.67–1.69)0.53 (0.28–1.00)
 Carotid ultrasonography (2013–2017)§0.89 (0.56–1.40)0.59 (0.33–1.06)
 MRI (2003–2007)§1.97 (1.07–3.60)5.25 (1.19–23.15)
 MRI (2008–2012)§2.20 (1.25–3.88)4.64 (1.09–19.80)
 MRI (2013–2017)§1.99 (1.14–3.49)5.21 (1.24–21.84)
 SPECT (2003–2007)§1.61 (1.01–2.58)1.42 (0.72–2.78)
 SPECT (2008–2012)§1.21 (0.77–1.90)1.19 (0.63–2.24)
 SPECT (2013–2017)§1.24 (0.80–1.92)0.65 (0.34–1.24)
  • Note: CI = confidence interval, CT = computed tomography, MRI = magnetic resonance imaging, SPECT = single-photon emission computed tomography.

  • * Estimates are for the 3-month look-back period; Appendix 3 (Supplementary Tables 5–6; available at www.cmajopen.ca/content/8/4/E695/suppl/DC1) provides estimates for each look-back period.

  • Total incidentally detected cases shown to obscure back-calculation of positron emission tomography–associated cases, which were excluded from imaging-specific models because of few cases.

  • Denotes a statistical interaction term among the covariates.

  • § Denotes the statistical interaction estimate among the covariate categories.