Black patients referred to a lung cancer screening program experience lower rates of screening and longer time to follow-up

Table 3 Results of Multilevel Analysis of Having LDCT Among Those Living in Philadelphia Planning Districts (n = 530)

	Model 1 Odds Ratio (95% Confidence Interval)	Model 2^a Odds Ratio (95% Confidence Interval)	Model 3^a Odds Ratio (95% Confidence Interval)	Model 4^a Odds Ratio (95% Confidence Interval)	Model 5^a Odds Ratio (95% Confidence Interval)
Individual Level
Race
• Black	0.44 (0.30–0.65)**	0.50 (0.32–0.77)**	0.50 (0.30–0.82)**	N/A
• White	1.00	1.00	1.00
Neighborhood Level
Neighborhood Disadvantage^b		0.53 (0.26–1.09)+
%blacks^c
low (2.5–24.2%)			1.00	1.00
high (39.5–93.4%)			0.82 (0.50–1.33)	0.55 (0.37–0.83)**
Race x %blacks
whites/low %blacks					1.00
white/high %blacks					0.99 (0.37–2.63)
blacks/low%blacks					0.54 (0.29–0.99)*
blacks/high%blacks					0.41 (0.26–0.65)**
Log Likelihood	− 314.75	− 283.58	− 294.19	− 293.84	−290.03

Note. +p < .10; *p < .05; **p < .01
^a Adjusted for known lung cancer risk factors: age, gender, BMI, COPD, family history of lung cancer, and smoking status
^b Composite measure of neighborhood disadvantage (%of adults without health insurance, %of population living in a household with an income below 100% of the Federal Poverty Line, %of unemployment, and %of less than some college)
^c Median split of %blacks (low vs. high)

ISSN: 1471-2407