• Presentation

Evaluating Methods for Increasing Physician Survey Cooperation

Citation

Olmsted, M., Murphy, J. J., McFarlane, E. S., & Hill, C. A. (2005, May). Evaluating Methods for Increasing Physician Survey Cooperation. Presented at American Association for Public Opinion Research 60th Annual Conference, Miami Beach, FL.

Abstract

Conducting surveys with physicians is different from conducting surveys in the general population. Physicians have demanding work schedules and participating in a survey is often not seen as a priority. They are frequently approached about taking part in surveys or other research, which can make them even more reluctant to participate. When confronted with a survey, many will refuse to respond, while others will agree to participate only after multiple persuasion attempts. In addition, physicians typically have a number of "gatekeepers," such as receptionists, administrative staff, or nurses who protect them from unwanted intrusions.

Over the years, a number of effective methods have been developed to increase the level of survey participation. These methods include pre-notification letters, incentives, reminders, survey form design, endorsement letters, sending additional surveys, shortening survey length, among others. Each of these methods has been demonstrated to have a positive impact on the level of participation in surveys with the general population. However, the success of many of these methods has not been adequately tested with special or elite populations such as physicians who typically have low rates of survey participation.

A key question for researchers considering methodological improvements is the cost to benefit ratio. Essentially, the question is which method will yield the best response at the lowest cost. To address this issue, researchers at RTI International conducted several experiments within a national survey of board certified physicians conducted for U.S. News & World Report. This paper assesses the cost and utility of differential reminders (post cards vs. 1st class letters) and length of survey (short vs. long forms) on survey response.

The presentation will describe relevant literature, cover the findings from the experiments embedded within this year’s study, and discuss possible future directions in survey research with physicians.