In recent studies of 'low natural killer (NK) cell syndrome', low NK activity was measured in individuals who were symptomatic, and therefore a causal relationship between low NK activity and infectious or other disease manifestations could not be concluded. However, preliminary work by members of our collaborative team (S.L. and R.H.) provided some indications for chronic low NK activity preceding and predicting subsequent infectious morbidity. This present study was designed to address this causal question in a larger sample, using a longitudinal design. Subjects were 106 healthy normal volunteers from the community. They were examined medically and psychosocially at baseline, and were then followed over a 6-month interval, with serial monthly assessments over the study period. The results supported our hypothesis that individuals who were currently healthy, but who exhibited a pattern of natural immunity characterized by persistently low NK cytotoxicity would be at risk for development of infectious sequelae over a 6-month follow-up period. The results also showed that younger age and the perception of more severe 'hassles' or stressors also predicted more infectious morbidity during the 6-month study period. Chronological age appeared to have both a direct, as well as indirect (via NK activity) association with illness outcome. Contrary to our expectation, the report of environmental stressors was directly associated with illness outcome, but not indirectly associated with outcome via natural immunity.