Aim: The study aimed to investigate whether a bystander’s emotional stress state affects dispatcher-assisted cardiopulmonary resuscitation (DA-CPR) in out-of-hospital cardiac arrest (OHCA). The primary outcome was initiation of chest compressions (Yes/No). Secondarily we analysed time until chest compressions were initiated and assessed how dispatchers instructed CPR. Method: The study was a retrospective, observational study of OHCA emergency calls from the Capital Region of Denmark. Recorded calls were evaluated by five observers using a pre-defined code catalogue regarding the variables wished investigated. Results: Included were 655 OHCA emergency calls, of which 211 callers were defined as emotionally stressed. When cardiac arrest was recognized, chest compressions were initiated in, respectively, 76.8% of cases with an emotionally stressed caller and 73.9% in cases with a not emotionally stressed caller (2.18 (0.80–7.64)). Cases with an emotionally stressed caller had a longer time until chest compressions were initiated compared to cases with a not emotionally stressed caller, however non-significant (164 s. vs. 146 s.; P = 0.145). The dispatchers were significantly more likely to be encouraging and motivating, and to instruct on speed and depth of chest compressions in cases with an emotionally stressed caller compared to cases with a not emotionally stressed caller (1.64 (1.07–2.56); 1.78 (1.13–2.88)). Barriers to CPR were significantly more often reported in cases with an emotionally stressed caller compared to cases with a not emotionally stressed caller (1.83 (1.32–2.56)). Conclusion: There was no significant difference in initiation of chest compressions or in time until initiation of chest compressions in the two groups. However, the dispatchers were overall more encouraging and motivating, and likely to instruct on speed and depth of chest compressions when the caller was emotionally stressed. Furthermore, barriers to CPR were more often reported in cases with an emotionally stressed caller compared to cases with a not emotionally stressed caller. Trial registration: We applied for ethical approval from The Danish National Committee on Health Research Ethics, but formal approval was waived. We received permission for storage of data and to use these for research of OHCAs in the Capital Region of Denmark by Danish Data Protection Agency (P-2021-670) and Danish Health Authorities (R-2,005,114). The study is registered at ClinicalTrials (NTC05113706).