BACKGROUND: Adverse drug reactions (ADRs) are known to cause hospitalisation, longer hospital stays, as well as higher healthcare costs and mortality. Unrecognised ADRs are anticipated throughout the medicine lifecycle as, before the medicine reaches the market, clinical trials are conducted for a short period on a limited number of people, who might underrepresent the actual population. After the medicine reaches the market, emergent information that could affect its benefit-to-risk balance is usually shared by regulatory agencies and pharmaceutical companies through medicine risk communications. Medicines risk communications aim to prevent harm to patients by targeting their behaviour, knowledge, and attitudes, as well as those of health care professionals (HCPs). Despite their important role in translating these communications into their clinical practice, HCPs do not always adhere to the recommendations provided in risk communications. Measurement of medicine risk communications' effectiveness does not necessarily guarantee their implementation, cost-effectiveness, or transferability in real-world situations. To enhance the impact of drug regulatory interventions, implementation science has been encouraged. However, implementation science was not previously used to identify factors affecting HCPs' implementation of medicines risk communications. A recently widely used framework is the Theoretical Domain Framework (TDF). In this systematic review, the TDF was employed to categorise a range of different factors that could affect HCPs' implementation of medicine risk communications within their clinical contexts. METHODS: The search strategy involved a set of predefined search terms and fifteen databases, such as EMBASE, PubMed, Web of Science and CINAHL PLUS. Searches were conducted from April to May 2018 and updated in June 2021 using PubMed, Scopus, and CINAHL PLUS. A second reviewer independently conducted the screening process of the initial search. The total number of records screened was 10,475. A study was included if it reported any factors influencing HCPs' uptake of medicine risk communications. Only studies with English or Arabic abstracts were included. Those studies that did not include pharmacovigilance-related medicine risk communications were excluded. Additionally, studies only assessing HCPs' practice or evaluating the effectiveness of risk minimisation measures were excluded. Likewise, studies related to occupational hazards, case reports, interventional studies, and studies not involving HCPs were excluded. In case the published information was insufficient to decide whether to include or exclude a study, the authors were contacted. Furthermore, the authors of seven eligible abstracts were contacted for full-text articles. The mixed method appraisal tool (MMAT) was used to evaluate the quality of the included studies. All included studies were assessed by one reviewer, and a total of 16 studies were assessed by two reviewers independently. Disagreements were resolved through discussion. Using thematic analysis and concept mapping, a narrative synthesis was performed, followed by a critical reflection on the synthesis process. This review presents the results of the concept mapping, which involved matching the identified factors to the TDF. RESULTS: A total of 28 studies were included. Eleven domains influenced HCPs' implementation of medicine risk communications. A large number of studies included factors related to the "Knowledge" domain (n = 23), followed by "Beliefs about Consequences" (n = 13), "Memory, Attention and Decision Processes" (n = 12) and "Environmental Context and Resources" domains (n = 12). Seven studies reported "social influences" and six studies included factors relating to "Goals", followed by four studies involving factors related to "Social/Professional Role and Identity". Underrepresented domains included "Emotion" (n = 2), "Beliefs about Capabilities" (n = 2), "Behavioural Regulation" (n = 1), and "Reinforcement" (n = 1). On the other hand, none of the identified factors were related to the "Skills", "Optimism", or "Intentions" domains. Except for "Beliefs about Consequences", most studies contributing to the other three most commonly reported domains ("Knowledge"; "Environmental Context and Resources"; and "Memory, Attention and Decision Processes") scored low (1 or 2 out of 5) on the MMAT quality assessment. Moreover, the same number of studies (n = 5) contributing to the "Beliefs about Consequences" domain had low (1 or 2 out of 5), and intermediate (3 out of 5) scores on the MMAT. CONCLUSION: Medicines risk communications are important tools for disseminating information that may influence the benefit-to-risk balance of medicines. Even though HCPs are required to implement the recommendations of these communications, they do not always adhere to them. Using the TDF enabled the categorization of the range of factors that affect whether or not HCPs implement the recommendations provided in a medicine risk communication. However, most of these factors relate to four domains only ("Knowledge"; "Beliefs about Consequences"; "Memory, Attention and Decision Processes"; and "Environmental Context and Resources"). Additionally, most of the studies contributing to three of these four domains were of low quality. Future research should focus on using implementation science to identify target behaviours for actionable medicine risk communications. Regulators should use such science to develop cost-effective strategies for improving the implementation of medicines risk communication by HCPs.