Eighty adult patients, 33 men and 47 women, mean age 46 years (SD 11.8, range 19-74 years), were evaluated 5 years after low-back surgery. The mean duration of symptoms before operation was 8.7 (SD 7.1) years. The purpose was to evaluate the 5-year outcome of low-back surgery, to find the best predictors for the outcome, and to find out if a correlation exists between the patient's sense of coherence and the outcome of low-back surgery. The mean Oswestry pain index for the whole group of patients improved from 3.8 to 2.7 (P < 0.001). The greatest improvement in pain was found in the group aged 35-50 years. In those over 50 years old, pain improved significantly more in women than men. Regarding walking ability, the mean Oswestry gait index for the whole group improved from 3.0 to 1.9 (P < 0.001), with men over 50 years old achieving the greatest improvement in their walking. The mean Oswestry total index for the whole group was 41% before surgery, reflecting severe disability, and 25% at follow-up, reflecting moderate disability (P < 0.001). There was no difference between the mean values for men and women. Patients who had undergone several previous operations fared less well in the Oswestry total index, though their improvement was still significant (P < 0.05). The postoperative Oswestry total index values correlated significantly with the sense of coherence (SOC) scale values (r = -0.23, P < 0.05). In all patients, the Oswestry total index before the index operation is suggested to be a predictor of the final outcome. In multiple regression analysis, the number of previous operations and the preoperatively recorded Oswestry total index appeared to be the best predictors for outcome of low-back surgery. We also found that the SOC scale correlated significantly with the Oswestry total index and seems to provide a possible explanation of ability to cope with the disability and pain associated with low-back disorders.