Table 3.

Estimated effects of pulmonologist visit on non-small cell lung cancer treatment and outcomes

	OR or HR Associated with Pulmonologist Visit	95% CI
Treatment, early-stage NSCLC (N = 5,488)
Single-stage logistic regression	1.26	1.11–1.45*
2SRI estimate	1.52	0.96–2.40
Residual	0.91	0.73–1.14
Survival, early-stage NSCLC
Single-stage Cox PH regression	0.80	0.75–0.85*
2SRI estimate	0.79	0.63–0.99†
Residual	1.00	0.90–1.12
Treatment, advanced-stage NSCLC (N = 6,426)
Single stage logistic regression	1.88	1.67–2.10*
2SRI estimate	1.44	0.78–2.66
Residual	1.14	0.85–1.52
Survival, advanced-stage NSCLC
Single-stage Cox PH regression	0.66	0.63–0.70*
2SRI estimate	0.89	0.68–1.17
Residual	0.87	0.76–0.99†

Definition of abbreviations: 2SRI = two-stage residual inclusion; CI = confidence interval; HR = hazard ratio; NSCLC = non-small cell lung cancer; OR = odds ratio; PH = proportional hazard; SEER = Surveillance, Epidemiology, and End Results.

Adjusted for duration of chronic obstructive pulmonary disease diagnosis, tumor characteristics, basic demographics, and health status. Data from SEER–Medicare NSCLC cases diagnosed between 2002–2005 (37).

^*Estimated OR or HR significant at P < 0.001.

^†Estimated OR or HR significant at P < 0.05.