Objectives. Longitudinal measurements can provide important information regarding variations in developmental trajectories of patients in long-term treatment. The present study investigated changes in general symptoms, depression, anxiety, and interpersonal problems during the first 2 years of long-term psychoanalytic psychotherapy (PP) and psychoanalysis (PA). It was expected that interpersonal problems would diminish more slowly compared to symptomatic dysfunction. Design. An accelerated longitudinal design with five consecutive measurement points across two cohorts of patients was used. Methods. Changes on the Symptom Checklist-90-R (SCL-90-R), Beck Depression Inventory-II (BDI-II), State-Trait Anxiety Inventory (STAI), and Inventory of Interpersonal Problems-64 (IIP-64) were investigated during the first 2 years of long-term PP (n= 73) and PA (n= 40). Linear regression analysis was performed to model the different courses of improvement. Results. After 2 years of treatment, patients in both groups still presented moderate to high levels of symptoms and interpersonal problems compared to non-clinical populations. As expected, interpersonal problems changed less rapidly. PP patients changed both with regard to symptomatic and interpersonal problems, whereas the only significant change in the PA group was on one of the symptomatic subscales. Slopes in the PA group and in PP group did not differ significantly from each other, except for the IIP-64 scale intrusive, with PP patients showing significantly more improvement than PA patients. The height of intake values of the outcome variables appeared to predict the speed of symptomatic recovery. Conclusions. Symptoms and interpersonal problems did not decrease notably within the first 2 years of psychoanalytic treatment. This is consistent with the idea that significant change takes time for patients with chronic mental disorders and personality pathology. In regular practice, it is advisable to monitor changes routinely in order to identify slow responders more quickly and change the treatment plan, if necessary. © 2011 The British Psychological Society.