Experience with nearly 300 applicants for predictive testing for Huntington's disease has shown that apart from the expected problems such as those related to third-party interests and the effects of an adverse test result, there were several less foreseen difficulties. These included the finding that some applicants were already clinically affected, requests for testing of minors, unintentional risk alteration for relatives, and the use of research samples for service purposes. More of the problems involved clinical and counselling aspects rather than laboratory procedures.