Table 3. Association between cervical cancer screening history and being diagnosed with CIS*.
| Variables | Matched | Un-matched | |||
|---|---|---|---|---|---|
| OR‡ | 95% CI | OR‡ | 95% CI | ||
| Overall | 2.40 | 2.18–2.65 | 2.43 | 2.25–2.63 | |
| Age at cancer diagnosis (yr)† | |||||
| 30–39 | 1.51 | 1.29–1.76 | 1.42 | 1.24–1.63 | |
| 40–49 | 2.28 | 1.91–2.71 | 2.27 | 1.99–2.59 | |
| 50–59 | 4.05 | 3.07–5.33 | 3.70 | 3.03–4.51 | |
| 60–69 | 4.78 | 3.15–7.23 | 4.64 | 3.45–6.24 | |
| 70–79 | 5.62 | 3.53–8.94 | 7.31 | 5.02–10.65 | |
| ≥80 | 6.86 | 3.10–15.15 | 4.99 | 2.90–8.58 | |
| Socioeconomic status† | |||||
| NHIS (premium >50%) | 1.90 | 1.62–2.22 | 2.03 | 1.79–2.30 | |
| NHIS (premium ≤50%) | 2.68 | 2.36–3.05 | 2.64 | 2.39–2.92 | |
| MAP recipients | 4.79 | 2.69–8.51 | 4.37 | 2.86–6.67 | |
| Screening frequency | |||||
| Never-screened | 1.00 | Reference | 1.00 | Reference | |
| Once | 1.65 | 1.45–1.87 | 1.78 | 1.63–1.94 | |
| Twice | 2.87 | 2.33–3.54 | 2.60 | 2.35–2.88 | |
| Three times or more | 5.10 | 4.03–6.45 | 3.92 | 3.55–4.33 | |
| p for trend | <0.01 | <0.01 | |||
| Time from last screening to cancer diagnosis | |||||
| Never-screened | 1.00 | Reference | 1.00 | Reference | |
| ≤11 mo | 2.51 | 2.24–2.82 | 2.77 | 2.55–3.00 | |
| 12–23 mo | 2.89 | 2.16–3.85 | 2.26 | 1.99–2.58 | |
| 24–35 mo | 2.11 | 1.43–3.10 | 2.05 | 1.74–2.42 | |
| ≥36 mo | 1.55 | 1.15–2.08 | 1.34 | 1.19–1.52 | |
| p for trend | <0.01 | <0.01 | |||
CI, confidence interval; CIS, carcinoma in situ; MAP, Medical Aid Program; NHIS, National Health Insurance Service; OR, odds ratio.
*Conditional and unconditional logistic regression models were used for the matched and un-matched datasets, respectively; †OR of detecting CIS in the ever-screened vs. never-screened groups in a subgroup; ‡ORs compare the odds of being diagnosed with CIS in ever-screened group to the odds of being diagnosed with invasive cervical cancer in ever-screened group.