INTRODUCTION: There is a wide variety in prescribing by primary healthcare workers in low-income and middle-income countries. While there is much information available, both online and offline, there is variation in quality and relevance to different settings. Acting on incorrect or out-of-date information can lead to inappropriate prescribing and impact on patient safety. The aim of this review is to systematically review the evidence on how primary healthcare workers obtain information during consultations to prescribe safely and appropriately. METHODS AND ANALYSIS: We will identify relevant articles by searching electronic databases: Medline (Ovid), EMBASE (Ovid), Cochrane Central Register of Controlled Trials, CABI Global Health (Ovid), WHO global health library, POPLINE, Africa-Wide Information (Ebsco), Library, Information Science & Technology Abstracts (Ebsco), ClinicalTrials.gov and WHO International Clinical Trials Registry Platform. Also, the Health Information For All network will be consulted and evidence databases (TRIP database, Epistemonikos, PDQ Evidence) will be searched. We will hand-search reference lists, run citation searches of included studies and email authors of identified papers. Observational and intervention studies involving primary healthcare workers in low-income and middle-income countries who prescribe and/or dispense medication will be included. The primary outcome is the proportion of healthcare workers obtaining information relevant to consultations from different sources. Secondary outcomes are the change in healthcare provider and patient knowledge or behaviour, adverse outcomes and use of resources. We will exclude studies focusing on secondary care. We anticipate a limited scope for meta-analysis and will provide a narrative overview of findings and tabular summaries of extracted data. ETHICS AND DISSEMINATION: No ethics approval is required. Findings will be disseminated through the Healthcare Information For All network. PROSPERO REGISTRATION NUMBER: CRD42018091088.