The aim of this research is to contribute knowledge to a scarce number of studies on negative effects (NE) in psychotherapy in order to increase awareness of both clinicians and patients along with effectiveness and safety of therapeutic approaches. Ninety-two participants took part in the study by non-random selection. Participants formed a purposive sample and an additional expert sample consisting of people with experience of both psychotherapy and NE. The design is mixed methods including between-groups comparisons using statistical tools of descriptive, non-parametric and regression testing plus short semi-structured interview. Two main measurements on NE and alliance were used: Negative Effects Questionnaire (NEQ) and Agnew Relationship Measure – 5 (ARM-5) by online survey. The results demonstrated that humanistic therapy shows significantly lower level of NE to psychoanalysis which led to the rejection of hypothesis. This is partly consistent with the body of research. CBT almost achieved significance to humanistic, however more research is needed. The childhood-past focus, number of sessions and gender of therapist showed no significant contribution to the development of NE, thus the hypotheses related were accepted. The result did not match the previous studies on those components. The next accepted hypothesis related to therapeutic alliance where the result demonstrated that negative effects decrease when alliance increases confirming a body of research indicating an importance of therapeutic relationship in psychological treatments. The qualitative results pointed to the need of formal explanation of NE to the patients and applying risk prevention plan at the beginning of treatment.