Effectively recruiting a representative patient sample is a critical step in health services and clinical research. In a common recruitment protocol, patients are first identified administratively as potential subjects, mail contact is made (giving patients the option to proactively opt-in or opt-out by calling the Data Collection Center) and additional telephone screening is conducted. This process not only is resource intensive, but can introduce bias due to telephone noncontact in an era of increased caller ID and call screening. A seemingly attractive alternative is the use of the internet for study screening and recruitment. However, this method is hamstrung by the lack of access to patient email addresses. In an attempt to overcome this hurdle and explore the utility of the internet for study recruitment, the BRINK (BRain IN Kidney Disease) Memory Study embedded a weblink in the initial postal mail contact to potential participants. This enabled us to observe the preference for web recruitment compared to inbound and outbound telephone contact. We also compared the demographic composition of those selecting each of the three modes. In total, we recruited 153 patients. Of these, 23 (15%) completed screening at the study website, 107 (70%) via inbound calls and 23 (15%) via outbound calls. While the average age of the recruited sample did not differ across modes, the gender distribution of those recruited through internet screening more closely matched the underlying population distribution. Given that internet recruitment requires significantly less resources, and given its positive impact on sample representativeness, this practice holds promise for future studies.