Table 4.

Patient- and provider-level predictors of screening completion in Asian patients by cancer screening type

Cervical cancer screening (N = 39,845)Mammography screening (N = 11,268)Colorectal cancer screening (N = 20,793)
naOR (95% CI)bnaOR (95% CI)bnaOR (95% CI)b
Patient characteristics
Sex
 Male9,3631.00
 Female11,4300.88 (0.83–0.95)
Age (per year)1.03 (1.02–1.03)0.99 (0.98–0.99)1.04 (1.04–1.05)
Race/Ethnicity
 Chinese13,5531.005,1961.009,4921.00
 Asian Indian16,8260.66 (0.62–0.71)2,0300.53 (0.47–0.60)4,6930.37 (0.34–0.40)
 Filipino4,7800.96 (0.84–1.10)2,0821.27 (1.07–1.50)3,3090.96 (0.85–1.08)
 Japanese1,6800.91 (0.83–0.99)1,1140.72 (0.64–0.82)1,8330.60 (0.55–0.66)
 Korean1,4710.88 (0.77–1.01)3390.78 (0.60–1.02)5540.92 (0.76–1.11)
 Vietnamese1,0451.14 (0.96–1.34)3091.54 (1.12–2.12)5500.99 (0.82–1.21)
 Native Hawaiian, Pacific Islander4900.57 (0.46–0.70)1980.53 (0.39–0.74)3620.48 (0.38–0.60)
Enrolled in “My health online”
 No25,1891.007,5301.0013,8071.00
 Yes14,6561.31 (1.24–1.37)3,7381.32 (1.20–1.46)6,9861.04 (0.98–1.11)
Primary language
 English34,2111.009,1161.0017,2011.00
 Not English, physician–patient concordant1,0750.94 (0.79–1.12)3390.86 (0.64–1.15)7110.94 (0.77–1.15)
 Not English, physician–patient discordant4,5591.02 (0.94–1.10)1,7530.81 (0.71–0.92)2,8810.79 (0.72–0.87)
Patient requires an interpreter
 No39,2881.0010,8841.0020,1471.00
 Yes5570.81 (0.66–1.01)3840.97 (0.74–1.27)6460.79 (0.65–0.95)
PCP visits in past 2 y (per visit)1.13 (1.12–1.14)1.22 (1.20–1.25)20,7931.14 (1.12–1.15)
Provider characteristics
Sex
 Male5,8171.002,4171.008,7321.00
 Female34,0281.66 (1.51–1.82)8,8511.16 (1.00–1.34)12,0611.05 (0.91–1.21)
Degree
 DO or NP/PA2,9701.005231.008721.00
 MD36,8750.97 (0.83–1.13)10,7450.93 (0.69–1.24)19,9211.02 (0.76–1.36)
Specialty
 Family medicine22,1491.005,4701.0010,0871.00
 Internal medicine17,6961.09 (1.01–1.18)5,7981.05 (0.92–1.19)10,7061.23 (1.07–1.40)
Any Asian language capability
 No32,4301.009,4731.0017,4101.00
 Yes7,4150.95 (0.85–1.08)1,7950.96 (0.78–1.17)3,3830.84 (0.67–1.05)
ICC0.0140.0330.061
  • aSample sizes (count of patients) provided for categorical variables only.

  • bORs generated using hierarchical multivariable logistic models with random intercept by provider and fixed effects at the provider-level (sex, degree, specialty, Asian language capability) and patient-level (sex, age, detailed race/ethnicity, enrolled in “My health online,” primary language concordance with provider, use of interpreter, PCP visits in the past 2 years).