TABLE 3.

Adjusted odds ratios from multivariable nominal logistic regression for care gap type among women diagnosed with cervical cancer

	Screening test failure (N = 156) a	Diagnostic/treatment care gap (N = 111) a
	Comparison group for outcomes: Lack of a screening test (N = 231) a
	Odds Ratio (95% CI) b
Diagnosis year
2010–2012	Ref	Ref
2013–2014	1.42 (0.83–2.43)	1.66 (0.89–3.08)
Age at diagnosis (years)
21–29	3.58 (1.18–10.80)	3.43 (1.04–11.30)
30–39	Ref	Ref
40–49	0.77 (0.40–1.49)	0.73 (0.36–1.51)
50–59	0.48 (0.24–0.93)	0.34 (0.15–0.73)
60–65	0.38 (0.15–0.93)	0.35 (0.13–0.96)
Race/ethnicity
Non‐Hispanic White	Ref	Ref
Non‐Hispanic Black	2.64 (1.07–6.47)	2.85 (1.00–8.13)
Hispanic	1.65 (0.93–2.93)	1.55 (0.82–2.96)
Asian/Pacific Islander	2.67 (1.07–6.67)	2.75 (1.02–7.38)
Other/Unknown	0.62 (0.17–2.24)	0.34 (0.06–1.90)
SEER summary stage at diagnosis
Localized	Ref	Ref
Regional	0.19 (0.11–0.35)	0.14 (0.07–0.29)
Distant	0.50 (0.23–1.11)	0.15 (0.04–0.55)
Unknown	1.31 (0.25–6.79)	1.62 (0.24–11.10)
Histology
Squamous cell carcinoma	Ref	Ref
Adenocarcinoma	3.91 (2.31–6.61)	3.12 (1.73–5.62)
Other	0.42 (0.13–1.34)	0.38 (0.09–1.66)
Study site
Site 1	1.30 (0.65–2.58)	1.75 (0.83–3.67)
Site 2	Ref	Ref
Site 3	0.25 (0.11–0.59)	0.10 (0.03–0.32)
Site 4	0.35 (0.17–0.73)	0.30 (0.13–0.69)

Abbreviations: CI, confidence interval; Ref, reference group; SEER, Surveillance, Epidemiology, and End Results.

^aOne woman without a clear care gap and 6 women with missing results for Pap test (thus it was not possible to correctly assign their care gap) were excluded, leaving 498 women in this analysis.

^bModel adjusted for all variables shown in this table. We did not include insurance type in this analysis due to inability to obtain stable estimates for insurance type. Additionally, adjusting for insurance type did not lead to material changes of the results presented in this table.