white boxwhite box

Sample Design and Subject Selection

The study population consisted of two groups: (1) EMT-Ps who were initially credentialed by NREMT and voluntarily chose to reregister with the NREMT, and (2) those who did not reregister their NREMT-Paramedic credential after the initial two-year registration cycle and chose to remain only credentialed by the State. The pool of subjects who reregistered was restricted to those who voluntarily did so because they were thought to be similar to the non-registered cohort. EMT-Ps required to reregister for State certification would likely have different characteristics than those in voluntary and nonreregistered cohorts and were not included in the study population.

To control for the varying amounts of continuing education, and to help ensure consistency, three cohorts of subjects, initially certified 2, 4, and 6 years earlier, comprised of those who voluntarily reregistered and those who did not reregister, were selected. Cohorts beyond 6 years were not used due to industry turnover and the resources required to locate them.

The sampling plan was implemented by asking all NREMT-Ps during the March 2001 reregistration cycle (a biennial requirement) why they renewed their NREMT certification. Those who stated “It is a requirement to work in my State” or “My employer requires NREMT certification” were eliminated. This assisted in ensuring that the voluntary inclusion factor was upheld. Also included in the study population were all of the cohort NREMT-Ps who did not reregister after their first opportunity to do so. No EMT-Ps who reregistered after their first opportunity were included in the study (disallowing enrollment to those not reregistering during their second or third cycle). Names and addresses of NREMT-Ps were obtained from the National Registry of Emergency Medical Technicians in Columbus, Ohio. NREMT-P cohorts who voluntarily reregistered and those who did not reregister were sent a letter by mail to their last known addresses, requesting that they complete a multiple-choice questionnaire on-line. Members of the nonreregistrant cohort were required to be State-certified EMT-Ps. Competence was determined by administering an exam, details of which are explained in Appendix B. A $25 incentive payment was sent to those subjects who completed a three-part survey instrument consisting of decommissioned NREMT registration items, an achievement exam, and a demographic questionnaire. Following the main survey, a followup survey of nonresponders was performed to determine if there was a difference between those who responded to the request to enroll in the study and those who did not. No appreciable descriptive differences were found (see Appendix C).

Figure 1 shows the obtained distribution of the cohort populations. The deeply shaded boxes show the sampling plan and the lightly shaded boxes show the numbers of actual subjects enrolled in the study. The figure demonstrates the prospective estimate of the minimum number of subjects needed to enroll in the study in order to to detect at least a 10 percent difference in exam scores among the cohorts with a 95 percent level of certainty. Ultimately 1,110 EMT-Ps were successfully recruited and completed the exam over a 6-month period. The subjects were self-selected from a maximum population pool of 17,129.

Figure 1. Subject Population Distribution

Figure 1 shows the obtained distribution of the cohort populations.

Survey Instrument

The three-component survey instrument consisted of: (1) a 60-question exam comprised of decommissioned NREMT-P certification items, (2) a 30-question achievement examination designed by a panel of experts, and (3) the 12-question demographic questionnaire used in the NREMT LEADS study (Brown et al., 2002)

Data Collection

A Web-based survey was used to collect all data from the subjects. Subjects logged in using a unique identifier, consented to participate in the study, and verified inclusion criteria . Instructions were provided to the participants, and once they began the survey they were required to complete it in its entirety. Ten questions were displayed at a time, and participants were not permitted to move to the next frame without completing all questions in that frame. Participants could change responses at any time within the 10-question frame they were in, but were not permitted to go to previous frames. E-mail address, street address, and phone contact information were provided should the participants want to reach the principal investigator, the institutional review board, or the study administrator. Participants who lost their Internet connections while completing the survey questionnaire were advised to contact the study administrator, who assigned them a new unique identifier. The investigators were not permitted access to the personal information of the subjects. Only a study administrator had access to this information so that incentive payments could be made. The researchers released randomly selected cohorts of names until predetermined enrollment targets from the different cohorts had been reached. Once targets were reached, subjects received a notice when they attempted to log onto the Web site. The requirement to complete the survey in order to collect the participatory honorarium resulted in all surveys being completed; that is, there was no missing data.