Seventy-five patients with an early RA (disease duration between 2 and 12 months; 79% female; average age 49 years; 72% with "definite" RA) were followed in a prospective study over a period of 2 years. Within this 24 month observation period there was a drop in the average diagnostic ARA count from 5.0 to 4.2, in the ESR from 48 to 31 mm within the first hour and also in the number of affected joints from 13 to 9. Pain intensity and functional capacity showed a slight decrease in the average score. Despite a consistent antirheumatic therapy (72% on RID's after one year) there was a noticeable increase from 23 to 58% in the prevalence of patients with any erosive changes in the X-ray. Within the two years of this study one third of the patients employed at the onset had to quit their job. The number of patients retired due to RA rose from 0 to 23%. A comparison of two groups of patients (employed versus retired) revealed no significant differences in the initial examination with one exception: The patients eventually retired by the end of the study were significantly older with an average age of 51 versus 39 in the group of still employed patients. On the other hand, by the end of the study the patients remaining employed for the duration of the 2 years were significantly less active and also less severely diseased and handicapped. The number of patients with clinically relevant depression (BDI) or anxiety (STAI) did not change significantly over the 24 month period. The prevalence rates showed a maximum of 30 and 37% respectively. Both characteristics demonstrate a high intraindividual stability. The linear correlation between the depression scores at month 0 and 24 revealed an rtt of 0.52 (0.53 for anxiety). Despite substantial changes in the somatic and social state the extent of the mental suffering remained approximately the same for the whole group as well as intraindividually.