We investigated differences in lung cancer care and outcome between academic and community settings for all lung cancer patients diagnosed during 1973-1976 in New Hampshire and Vermont. Trained abstracters reviewed hospital charts to record personal, diagnostic, and clinical information, and survival was determined for all patients through the end of 1979. Patients diagnosed in university hospital cancer centers underwent more staging procedures and tended to be assigned to a higher stage than similar patients diagnosed in community hospitals. When tumor stage was considered as a covariable in a survival analysis, these patients appeared to have a lower mortality rate both for non-small cell tumors (mortality rate ratio, 95% confidence interval = 0.81, 0.71-0.91) and for small cell tumors (0.71, 0.55-0.91). When functional status rather than tumor stage was used to adjust for disease severity, there was no apparent survival advantage for university patients with non-small cell cancer (0.96, 0.85-1.09) and the lower mortality for small cell cancers (0.76, 0.59-0.97) was attenuated, although still statistically significant. We conclude that inconsistently-collected data on clinical stage can complicate comparisons of prognosis between cancer patients from different types of hospitals and that measures of performance status may be more useful indicators of disease severity in population based studies.