CDC Memo

Thank you for taking the time to share your concerns about our proposed survey to assess occupational exposure management practices. You raised three major points about response rate, sensitivity of information, and confidentiality protection for the data to be collected. Unfortunately, the project officer from the contractor who is supposed to be conducting the survey was unable to participate in the conference call last Thursday and is out all this week on paternity leave. His team lead, Marshall Ford has provided me with some of the information in my response to you.

1. Response rate. Although published literature about response rates to surveys state that 70% response rates are the minimum desired, this rate in general is applicable to the general population, e.g., household surveys. Healthcare providers are a unique target group and traditionally have much lower response rates to surveys. One paper, a review of physician response to surveys by S. Kellerman and J. Herold states “..low response rates to physician surveys are common..” Rates well below 70% (e.g., as low as say 28%, with standard deviations of say 19%).  

Although a low response rate may affect the validity of the data to be collected, CDC’s primary objective is to get a “snapshot” of the status of occupational exposure management in a variety of healthcare settings, since most of the limited information we have, is about acute-care settings (i.e., hospitals). We will use the collected information to assess the need for additional outreach, through the web, through print media, etc. to improve occupational exposure management in healthcare settings. The lack of representativeness will not be a major impediment to making decisions about developing ancillary materials, besides our guidelines.

Additional measures that could be incorporated in our procedures to increase response rates include sending reminder postcards, sending out more surveys, and emphasizing the importance of responses in the cover letter

2. Although the information we collect could be viewed as sensitive by the respondents, we don’t view it that way. Furthermore, CDC is not a regulatory agency. Information we collect will not be used to penalize respondents for lack of compliance with out recommendations. CDC, more specifically DHQP when it was known as the Hospital Infections Program, collected data about TB infection control measures in the early 1990s. The respondents did not all say that they were fully compliant with CDC guidelines. Perhaps a modification to the invitation/cover letters could emphasize that the survey is voluntary and that CDC is not a regulatory agency.

3. The confidentiality of data, especially the identity of the respondents is being handled by the contractor in the following manner. A unique identifier is associated with a survey. Once a response is received, that identifier is stripped from the data. The identifier is used to determine which facilities need additional follow-up as a reminder to complete the survey. Because there is no way to link a questionnaire with at specific respondent, it is not possible to determine compliance or lack of compliance by specific known facilities. Again, the cover letter(s), could emphasize this.

Adelisa (Lisa) Panlilio, MD, MPH
CAPT, USPHS
Medical Epidemiologist
Division of Healthcare Quality Promotion
Centers for Disease Control and Prevention
1600 Clifton Rd, NE Mailstop A-31
Atlanta, GA 30333
Phone (404) 639-4328
Fax (404) 639-4046
E-mail [email protected]