PURPOSE: To look for survival differences between men and women with limited stage small cell lung cancer (LS-SCLC) by examining stratified variables that impair treatment efficacy. METHODS: A retrospective review of 215 LS-SCLC patients treated from 1989 to 1999 with concurrent chemotherapy-radiotherapy modelled on the 'early-start' thoracic radiotherapy arm of a National Cancer Institute of Canada randomized trial. RESULTS: Of 215 LS-SCLC patients, 126 (58.6%) were men and 89 (41.4%) were women. Smoking status during treatment for 186 patients (86.5%) was: 107 (58%) nonsmoking (NS) (76 [71%] male [M]; 31 [29%] female [F]) and 79 (42%) smoking (S) (36 M [46%]; 43 F [54%]) (continuing-to-smoke F versus M, P=0.001). Fifty-six patients (26%) had radiotherapy interruptions (RTI) during chemotherapy-radiotherapy because of toxicity. Radiotherapy breaks were not associated with sex (P=0.95). Survival by sex and smoking status at two years was: F + NS = 38.7%; F + S = 21.6%; M + NS = 22.9%; and M + S = 9.1% (P=0.0046). Survival by sex and RTI status at two years was: F + no RTI = 32.4%; F + RTI = 23.6%; M + no RTI = 23.0%; and M + RTI = 3.8% (P=0.0025). Diffusion capacity for carbon monoxide (DLCO) was recorded for 86 patients (40%) and median survival by sex and DLCO was F = 16.7 months and M = 12.1 months for a DLCO less than 60%; and for a DLCO 60% or more, F = 15.1 months and M = 15.3 months. First relapses were recorded in 132 cases (61%), with chest failure in men (45%) greater than for women (35%) and cranial failure rates similar between sexes (48%). Upon multivariable analysis, continued smoking was the strongest negative factor affecting survival. CONCLUSIONS: In LS-SCLC, women overall do better than men, with or without a negative variable. The largest quantifiable improvement in survival for women came from smoking cessation, and for men from avoidance of breaks during treatment.