Chronic diseases such as diabetes and hypertension and unintentional injuries such as falls pose a significant burden on society, both in human suffering and financial cost (1–3). Most interventions in chronic disease prevention and treatment and in fall prevention are limited to people seen in a health care provider’s office, a clinic, or a hospital (4–7). Interventions are needed to reach high-risk people in the community who may or may not have such regular sources of care. Every community in the nation has an emergency medical services (EMS) system, and EMS providers see thousands of people each year in response to a 911 call. EMS providers collect information during patient encounters that may be indicators of chronic diseases such as high blood pressure or low blood glucose or injuries related to falls (8). Potential exists for educating high-risk EMS patients about chronic disease control and injury prevention during an EMS encounter (9). However, EMS providers, including emergency medical technicians (EMTs), firefighter EMTs, and paramedics, are acute-care providers whose main responsibility is to stabilize or transport a patient as efficiently and quickly as possible. Thus, public health interventions may not be perceived as compatible with the mission or duties of these providers.

Research shows that patient education materials alone can affect knowledge of chronic disease (10) and effect behavior change, particularly if these materials are part of an interpersonal intervention (11,12). In one study, EMS providers engaged older adults in a brief interpersonal intervention regarding heart attack symptoms during a nonemergency home visit and provided them with patient education materials. Results indicated that the combination of interpersonal intervention and patient education materials had a greater effect on patient behavior than patient education materials alone (11). Because nonemergency home visits are not a cost-effective way of reaching at-risk community residents with chronic disease information, we developed a patient education intervention to be delivered by EMS providers at the scene during an emergency patient encounter.

This article reports on the evaluation of a brief, at-scene patient-education intervention that included delivery of patient education materials by EMTs at part of an EMS response. The objective of our study was to assess the factors that influence adoption of the intervention by firefighter EMTs in fire departments in King County, Washington, a large metropolitan area in the Pacific Northwest. We used the Diffusion of Innovation theory to develop hypotheses regarding firefighters’ adoption of the at-scene intervention (13).

Diffusion theory suggests that an innovation can relate to any product, policy, idea, or practice that is different from the status quo. An at-scene educational intervention would likely be perceived as something new for EMS providers because the focus of their job is on acute patient care rather than education. Diffusion theory further postulates that adoption of an innovation (ie, patient education program) depends on 1) awareness of the program; 2) evaluation of program attributes (is the innovation perceived to be superior to the status quo [relative advantage], compatible with beliefs, costly, difficult or easy to adopt, observable [are others doing it]); and 3) communication with peers or superiors (social influence). One of the assumptions of diffusion theory is that adoption of an innovation is largely under the individual’s control as opposed to adoption of new clinical guidelines, which are mandated by the individual’s superior (13).

We received 822 survey responses from 2,047 firefighter EMTs in the county for an overall response rate of 40.1%. Response rates varied by fire department with a mean response rate of 43% (range, 8.0%–90.0%). We received no responses from 2 departments and therefore excluded them from analysis. Response rates were higher in departments that handed out paper versions of the survey than in departments that used the Web-based survey, a mean response rate of 45.0% versus 27.0%.

Survey respondents reflected the demographic characteristics of the King County firefighter EMT population as a whole: largely male (94%), non-Hispanic white (80%), and middle-aged (mean 41 years). Of the 822 responses, 19% reported they almost never handed out the MIRF backer to appropriate patients; 46%, sometimes; 23%, often; and 11%, almost always. Fifty-four percent of respondents reported that they often or almost always were able to discuss the information on the pamphlet when they handed it out.

Self-reported use of the MIRF backer varied under different circumstances (Table 1). The data show that EMTs were most likely to distribute the pamphlet to patients who were left at the scene. Circumstances in which the MIRF backer was less likely to be used included responses to 911 calls from nursing homes or care centers or in the presence of a language barrier.

EMTs who reported greater understanding of the MIRF-backer, its relative advantage, behavioral norms, perceived receptivity of the patient or family and friends to the pamphlet, and less difficulty remembering to give the pamphlet to the patient were more likely to report handing out the pamphlet often or almost always than EMTs who did not report regular use of the pamphlet during patient care (Table 2). In addition, encouragement from peers, training officers, and fire chiefs was significantly associated with adoption of the MIRF backer. Lack of encouragement of any kind was significantly negatively associated with handing out the backer. Age and sex were not significantly associated with MIRF-backer use (Table 3). Of the 822 survey respondents, 657 (80%) reported their race as white. Of the 20.0% (n = 164) who did not report their race as white, 73% (n = 119) did not report any race. Because of this high percentage of missing values, we did not include race in our analyses.

Perceptions of behavioral norms (“most EMTs in my department hand out the MIRF backer”) and relative advantage, remembering to hand out the MIRF backer, and encouragement were significantly associated with self-reported use of the MIRF backer in the multivariate analysis (Table 3).

We explored factors related to the adoption of an at-scene patient education intervention focused on prevention of chronic disease and unintentional injury and delivered by firefighter EMTs. The results show that strong predictors of use of the MIRF backer were perceived benefits (relative advantage) of using the MIRF backer during 911 incidents; encouragement from peers, training officers, and fire chiefs; and perceptions of behavioral norms on use of the MIRF in one’s own fire department.

Although age was not significantly associated with use of the MIRF backer, junior firefighter EMTs reported greater use of the MIRF backer. The fire departments in King County have participated in many other patient outreach programs over the years (11,15,16), and enthusiasm for new programs may wane over time. Alternatively, newly trained first responders may be more open to new approaches to patient care because they might not have firmly established habits. More research is needed to assess whether patient education programs would be more readily adopted if a patient education intervention like ours were part of EMT training of new employees.

Previous research shows that EMS-delivered chronic disease education can affect knowledge and behavior (11,12,16), but monetary costs of such interventions are high. Our study shows that a brief at-scene intervention in combination with patient materials is feasible. EMS providers see so many patients each year that reaching only a small percentage of them with the intervention could have a significant effect on patient health, particularly for community residents who do not have regular health care. Among the patients seen by the 29 fire departments, approximately 35% were not transported to hospital, representing approximately 56,000 patient interactions per year in the study area for whom the 911 encounter may be their only interaction with the health care system. The results of our survey suggest that these patients are perceived as the best targets for an intervention by EMS providers. On a community level, this will translate into a large number of residents receiving patient education at the scene who otherwise might not interact with the health care system.

Our study had limitations. Although the sample size was large, the overall response rate was low and varied dramatically among the fire departments. However, sufficient variation in the outcome measure (use of the MIRF backer) allowed for comparisons of adoption patterns between 2 groups of providers, those who handed out the pamphlet often or always versus those who handed it out almost never. We relied entirely on EMS providers’ self-reported use of the MIRF backer. Identification and timely follow-up with patients who received the MIRF backer was too challenging to be a feasible source for data collection. Future research needs to explore other avenues for confirming at-scene patient education practices.

New patient education initiatives in fire departments should be accompanied by clear and widespread communication of the purpose of the initiative, and EMS providers should be encouraged to adopt the intervention voluntarily. Following diffusion theory principles, it would be interesting to further investigate characteristics of individual fire departments (in addition to individual-level [EMT] characteristics) in relationship to the adoption of patient education interventions, including the presence of an advocate for the program, communication channels and networks, and early-adopter versus late-adopter EMT crews or fire departments.

The MIRF backer was perceived as more useful for patients who were left at scene rather than transported, indicating that the pamphlet was most useful in nonurgent EMT encounters. This group of patients is less likely to receive education than patients who are transported to hospital. As such it is encouraging to note that EMTs perceive this group of patients as appropriate for the intervention. Further investigation needs to be conducted on the effect of such interventions on patient knowledge and behavior.