A 3-page survey (Appendix 1) was developed to allow hospital pharmacy directors to comment on their institutions’ policies related to patients’ own medications. Respondents were also asked to provide copies of the policies and evaluations of the policies, if available. A similar survey administered to small US hospitals was obtained for guidance in developing the Canadian survey. The survey was pilot-tested by 3 hospital phar­macy practitioners not involved in the project, and improvements were made on the basis of their feedback.

Canadian hospitals were obtained on June 26, 2004, from the website of the Hospital Pharmacy in Canada), a survey of pharmacy directors of Canadian hospitals having at least 50 acute care beds. Electronic communication was chosen for this survey because of time and resource constraints. Ethics approval was waived by the local institutional research ethics board. The survey was sent by e-mail to 178 separate e-mail addresses in August 2004. The first author’s pharmacy director provided a cover message inviting recipients to participate. Twelve addresses were subsequently excluded: for 11 the message was returned as undeliverable, and the 12th was the institution where the first author (H.L.) was employed; this left a potential sample size of 166. A follow-up message was sent by e-mail 4 weeks later to increase the response rate. Pharmacists in Nova Scotia were also contacted individually by e-mail in an effort to obtain more local data from the first author’s home province. The individual e-mails generated 5 more survey responses. The survey was presented in English only. Most surveys were not returned anonymously; either the e-mail address, fax number, or full address of the respondent was included in the return survey by most respondents. kamagra tablets

Statistical Analysis

Responses were entered into a computerized database, which was used to gener­ate descriptive statistics.