A recent study demonstrated that patients treated with amputation and those treated with reconstruction had comparable functional outcomes at two years following limb-threatening trauma. The present study was designed to determine whether those outcomes improved after two years, and whether differences according to the type of treatment emerged. Three hundred and ninety-seven patients who had undergone amputation or reconstruction of the lower extremity were interviewed by telephone at an average of eighty-four months after the injury. Functional outcomes were assessed with use of the physical and psychosocial subscores of the Sickness Impact Profile (SIP) and were compared with similar scores obtained at twenty-four months. On the average, physical and psychosocial functioning deteriorated between twenty-four and eighty-four months after the injury. At eighty-four months, one-half of the patients had a physical SIP subscore of > or = 10 points, which is indicative of substantial disability, and only 34.5% had a score typical of a general population of similar age and gender. There were few significant differences in the outcomes according to the type of treatment, with two exceptions. Compared with patients treated with reconstruction for a tibial shaft fracture, those with only a severe soft-tissue injury of the leg were 3.1 times more likely to have a physical SIP subscore of 5 points (p < 0.05) and those treated with a through-the-knee amputation were 11.5 times more likely to have a physical subscore of 5 points (p < 0.05). There were no significant differences in the psychosocial outcomes according to treatment group. Patient characteristics that were significantly associated with poorer outcomes included older age, female gender, nonwhite race, lower education level, living in a poor household, current or previous smoking, low self-efficacy, poor self-reported health status before the injury, and involvement with the legal system in an effort to obtain disability payments. Except for age, predictors of poor outcome were similar at twenty-four and eighty-four months after the injury. The results confirm previous conclusions that reconstruction for the treatment of injuries below the distal part of the femur typically results in functional outcomes equivalent to those of amputation. Regardless of the treatment option, however, long-term functional outcomes are poor. Priority should be given to efforts to improve post-acute-care services that address secondary conditions that compromise optimal recovery.