BackgroundMedical assistance in dying (MAiD) traverses challenging and emotionally overwhelming territories: healthcare providers (HCPs) across jurisdictions experience myriad of affective responses secondary to possible tensions between normative and interwoven values, such as sanctity of life, dignity in death and dying and duty to care.ObjectiveTo determine the emotional impact on HCPs involved in MAiD.MethodsInclusion restricted to English language qualitative research studies from four databases (OVID Medline, EMBASE, CINAHL and Scopus), from beginning until 30 April 2021, and grey literature up to August 2021 were searched. Key author, citation and reference searches were undertaken. We excluded studies without rigorous qualitative research methodology. Included studies were critically appraised using the Joanna Briggs Institute’s critical appraisal tool. Analysis was conducted using thematic meta-synthesis. The cumulative evidence was assessed for confidence using the Confidence in the Evidence from Reviews of Qualitative Research approach.ResultsThe search identified 4522 papers. Data from 35 studies (393 physicians, 169 nurses, 53 social workers, 22 allied healthcare professionals) employing diverse qualitative research methodologies from five countries were coded and analysed. The thematic meta-synthesis showed three descriptive emotional themes: (1) polarised emotions including moral distress (n=153), (2) reflective emotions with MAiD as a ‘sense-making process’ (n=251), and (3) professional value-driven emotions (n=352).DiscussionThis research attempts to answer the question, ‘what it means at an emotional level’, for a MAiD practitioner. Legislation allowing MAiD for terminal illness only influences the emotional impact: MAiD practitioners under this essential criterion experience more polarised emotions, whereas those practising in jurisdictions with greater emphasis on allaying intolerable suffering experience more reflective emotions. MAiD practitioner’s professional values and their degree of engagement influence the emotional impact, which may help structure future support networks. English language literature restriction and absence of subgroup analyses limit the generalisability of results.