Objectives: Patient involvement in primary care research is often hindered by confidentiality concerns regarding the use of their identifiable medical records. Consequently there is no universal 'pre-consent' process in England, whereby patients provide prior informed consent for their identifiable data to be accessed for different research purposes. This mixed-methods two-phase study investigated patients' opinions about this proposed 'pre-consent' process and the effectiveness of different invitation methods. Methods: Phase 1 used cognitive and semi-structured interview methods to optimise recruitment documents for, and ascertain participants' attitudes towards, providing an informed consent decision for the 'pre-consent' process. Phase 2 tested the feasibility of different recruitment documents ('standard' and 'modified') and invitation methods (postal, consultation and new) to determine the most effective in terms of allowing patients the opportunity to provide an informed consent decision. Results: Phase 1: Readability of the 'standard' recruitment documents impacted on participants' abilities to provide consent decisions, consequently a 'modified' version was created using participant feedback. Opinions about the 'pre-consent' process were chiefly affected by an individual's personal attitudes; key findings were 'data security' concerns and 'public benefit' motives. Phase 2: Recruitment documents affected response rates, with patients that received 'standard' documents 43% less likely to re~pond than those receiving the 'modified' documents. Postal, consultation and new invitation methods biased the types of patients invited, the number and types of patients responding, and the consent levels provided. Of the 2550 patients invited to participate in the 'pre-consent' process, only 30% (n=767/2550) responded, undermining the feasibility of this process. Discussion: The 'pre-consent' process seems currently unfeasible. However, engaging with the public to identify the most effective recruitment documents, invitation methods and consent options could streamline research in primary care. These approaches, employed on a study-by-study basis, would ensure primary care research remains cost-effective and representative of the general population.