Study Sample

A packet consisting of a survey instrument, a cover letter, and a postage-paid return envelope without any financial or other incentive was mailed to a random sample of 690 physicians (625 excluding pediatricians) who were listed as members of the Minnesota Medical Association (MMA) (n = 460; 418 excluding pediatricians) or the North Dakota Medical Association (NDMA) (n = 230; 207 excluding pediatricians), respectively. We used a chi-square test for the equality of two proportions to determine the original sample size of 690.

We chose 690 by using the most conservative reference proportion (0.5 for one group) and an effect size of 0.10 in order to be able to detect a difference of 10% between the two proportions on primary variables of interest. We also used an alpha value of 0.05 and a power of 0.80, as per convention, and a two-sided test. The 2:1 (460:230) ratio was used mainly to ensure adequate representation from both states and was factored into the calculations for the chi-square test.
online pharmacy prescription drugs

Membership in the MMA and the NDMA included approximately 65% of practicing physicians in each state. Surveys were mailed between March 2006 and May 2006. Up to two additional surveys were mailed to non-responders. A follow-up postcard reminder was mailed approximately two weeks after the first and second survey mailings.

Approximately two weeks after the last survey mailing, non-responders were contacted by telephone and were sent an additional survey if they requested one. Phone surveys were conducted for physicians who indicated that preference.

The study received the approval of North Dakota State University’s institutional review board.

Survey Design

The survey was adapted from an instrument developed and used by Shrank et al. This modified instrument was pilot-tested on practicing physicians in Minnesota and North Dakota, and based on their responses, the survey was refined. The final survey consisted of 43 items, including a variety of questions on:

• physicians’ perceptions and beliefs about prescribing, drug costs, and out-of-pocket costs (i.e., out-of-pocket drug costs) in three-tier pharmacy benefit systems
• physicians’ personal awareness of patients’ out-of-pocket costs at the time of prescribing
• awareness of the determinants of those costs (i.e., insurers, formularies, pharmacy benefit structures)
• Medicare Part D

The survey consisted of multiple-choice, 5-point Likert-type scale, and open-ended questions. We also asked physicians for their comments. buy levitra uk

Data Analysis

Descriptive statistics were used to portray the characteristics of the respondents and to report the primary variables of interest of the study. The primary variables of interest were (1) physicians’ perceived responsibility for advising their senior patients about Medicare Part D enrollment and plan selection and (2) the percentage of senior patients they were advising about Medicare Part D enrollment and plan selection.

Other variables assessed (for potential associations with the primary variables of interest) were as follows: physicians’ access to necessary information about the patient’s medication regimen and annual drug costs; the content of formularies that would best serve the patient; and the least expensive plan for the patient.

We also assessed whether a physician’s specialty (generalist or specialist), formulary usage, or use of information technology when prescribing was associated with any of the aforementioned variables of interest. buy antibiotics canada

We performed chi-square analyses with Fisher’s Exact Test to assess these associations using SAS 9.1 (SAS Institute, Cary, NC, 2005). All bivariate analyses were based upon subgroup responses, not total response rate, because data imputation was not performed. Statistical significance was established at P = 0.05.