Skip to main content
. 2020 Dec 28;3(12):e2031349. doi: 10.1001/jamanetworkopen.2020.31349

Table 2. Multilevel Mixed-Effect Logistic Regression Models for AS or WW Compared With Active Treatment .

Variables OR (95% CI)
Model 1 (region only) Model 2 (model 1 + patient factors in fixed-effect term) Model 3 (model 2 + county factors in fixed-effect term)
Patient-level variables
Age, y NA
≤50 1 [Reference] 1 [Reference]
51-60 1.33 (1.21-1.46) 1.33 (1.21-1.46)
61-70 1.86 (1.70-2.04) 1.86 (1.70-2.04)
71-80 2.26 (2.04-2.50) 2.26 (2.05-2.50)
Race/ethnicity NA
Non-Hispanic White 1 [Reference] 1 [Reference]
Non-Hispanic Black 1.01 (0.95-1.07) 1.01 (0.95-1.07)
Hispanic 0.79 (0.74-0.84) 0.79 (0.74-0.85)
Asian/Pacific Islander 0.97 (0.88-1.07) 0.97 (0.88-1.07)
Other 0.94 (0.88-1.07) 0.94 (0.84-1.05)
Health insurance NA
Private and/or Medicare 1 [Reference] 1 [Reference]
Medicaid 0.73 (0.66-0.81) 0.73 (0.66-0.81)
Uninsured 1.07 (0.91-1.25) 1.07 (0.91-1.25)
Marital status NA
Married 1 [Reference] 1 [Reference]
Single 1.29 (1.23-1.34) 1.29 (1.23-1.35)
PSA, ng/mL NA 0.98 (0.96-0.99) 0.98 (0.97-0.99)
Clinical T category NA
T1c 1 [Reference] 1 [Reference]
T2a 0.79 (0.73-0.84) 0.79 (0.73-0.84)
% Biopsy-positive cores NA
<33% 1 [Reference] 1 [Reference]
≥33%, <50% 0.46 (0.43-0.49) 0.46 (0.43-0.49)
≥50% 0.29 (0.27-0.32) 0.29 (0.27-0.32)
Year of diagnosis NA
2010 1 [Reference] 1 [Reference]
2011 1.31 (1.23-1.40) 1.31 (1.23-1.40)
2012 1.89 (1.77-2.01) 1.89 (1.77-2.01)
2013 2.87 (2.69-3.06) 2.87 (2.69-3.06)
2014 3.24 (3.03-3.47) 3.24 (3.03-3.47)
2015 3.58 (3.35-3.83) 3.58 (3.35-3.83)
County-level variables
City type, population NA NA
Metropolitan, >1 million 1 [Reference]
Metropolitan, 250k-1 million 0.89 (0.71-1.11)
Metropolitan, <250k 0.76 (0.59-0.98)
Urban, >20k 0.89 (0.67-1.17)
Urban, 2500-19 999 0.82 (0.63-1.06)
Rural/urban, <2500 0.97 (0.68-1.37)
Median household income, $ NA NA
<50k 1 [Reference]
≥50k, <75k 1.11 (0.91-1.37)
≥75k, <100k 1.14 (1.14-1.14)
≥100k 1.09 (0.70-1.71)
Median No. of ≥ college education per 100k NA NA
<15k 1 [Reference]
≥15k, <30k 1.19 (0.96-1.48)
≥30k, <45k 1.27 (0.91-1.76)
≥45k 1.29 (0.76-2.18)
Median No. of urologists per 100k NA NA
0 1 [Reference]
>0, ≤2 0.98 (0.75-1.28)
>2, ≤4 0.99 (0.78-1.27)
>4 ≤6 0.84 (0.62-1.13)
>6 0.88 (0.62-1.27)
Median No. of radiation oncologists per 100k NA NA
0 1 [Reference]
>0, ≤1 0.93 (0.69-1.26)
>1, ≤2 1.07 (0.84-1.37)
>2 ≤3 1.04 (0.76-1.42)
>3 0.90 (0.64-1.27)
Median No. of PCPs per 100k NA NA
0 1 [Reference]
>0, ≤50 0.93 (0.49-1.76)
>50, ≤100 0.93 (0.48-1.80)
>100 1.07 (0.52-2.20)
Median No. of hospital beds per 100k NA NA
0 1 [Reference]
>0, ≤200 1.20 (0.93-1.56)
>200, ≤400 1.15 (0.87-1.51)
>400, ≤600 1.16 (0.83-1.63)
>600 1.05 (0.75-1.48)
Random-effects of regional variables (SEER registry/county)
Variance of random effect (SEER registry) 0.40 0.47 0.34
Variance of random effect (county) 0.32 0.34 0.32
Intraclass correlation 0.18 0.19 0.17
Median odds ratio 2.25 2.37 2.17

Abbreviations: NA, not applicable; OR, odds ratio; PCP, primary care physician; PSA, prostate-specific antigen; SEER, Surveillance, Epidemiology, and End Results.