A survey of EMS directors was developed in 2006 with input from MSP members to collect information about prehospital stroke care capacity and practices among Minnesota ambulance services. The survey also provided information to MSP about stroke training needs and effective methods to train and educate EMS personnel.

A sample of ambulance directors beta-tested the 23-question survey. The MDH Institutional Review Board approved the final survey tool in May 2006. The list of recipient ambulance services was obtained from the Minnesota Emergency Medical Services Regulatory Board. EMS agencies providing only air transport or facility-to-facility transport services were excluded because this survey focused on ground transport issues.

In June 2006, the survey tool, along with a cover letter written jointly by the MSP and MDH, was sent to the 240 directors overseeing all 257 ambulance services in the state. Respondents were given the option to complete the survey online. Two follow-up letters were sent to ambulance services not responding within 6 weeks. A total of 199 (77% response rate) surveys were completed. The survey closed in August 2006. A copy of the survey is available at http://www.health.state.mn.us/cvh.

MSP members with expertise in clinical stroke care developed the survey of hospitals. Questions were based on recommendations from the Brain Attack Coalition for stroke centers (5) and from the American Stroke Association Stroke Systems of Care (4) planning committee for Minnesota. The MDH Institutional Review Board approved the final survey of 22 questions in October 2005. The list of recipient hospitals was obtained from the MDH Facility and Provider Compliance Division Web site (www.health.state.mn.us/divs/fpc/fpc.html). The survey excluded hospitals that serve primarily children or otherwise are specialty care facilities (e.g., rehabilitation hospitals) because this assessment focused on adults receiving acute care.

Every hospital was contacted by telephone to determine the best recipient of the survey. Recipients were mostly directors of nursing or quality improvement. The survey, along with a cover letter written jointly by MSP and MDH, was sent to 133 acute care hospitals in January 2006. Respondents were given the option to complete the survey online. Two follow-up letters were sent to hospitals not responding within 6 weeks. Hospitals that did not return the second follow-up survey were contacted by telephone to conduct the survey. A total of 120 (90%) of 133 hospitals responded. Nineteen hospitals within the seven-county Minneapolis–St. Paul metropolitan area ("Twin Cities metro area") responded, compared with 101 rural (non-metro and small metro area) hospitals. Of the 13 hospitals not responding to the survey, 11 reported fewer than 100 beds, and 11 are located outside the Twin Cities metro area. The survey closed in April 2006. A copy of the survey is available at www.health.state.mn.us/cvh.

Data for both surveys were analyzed by size of hospital (according to number of licensed beds) and location (Twin Cities metro and rural). Hospitals were grouped into the following categories: small (<50 beds [n = 75]), medium (50–99 beds [n = 17] and 100–249 beds [n = 9]), and large (≥250 beds [n = 19]).

Stroke patient transportation and advance notification

Fewer than two-fifths (37%) of respondents reported the existence of a written transportation protocol for stroke patients (Table 1). Most (91%) would not bypass a hospital to send a patient to a destination with greater capacity for stroke care; according to written comments, ambulances take patients to a hospital chosen by the patient or patient's family. Most ambulances (78%) provide advance notice to hospitals about stroke patients needing urgent care, with notification on scene (10%), immediately en route (21%), or as soon as possible en route (54%); the questionnaire did not define the difference between "immediately" and "as soon as possible." Ten percent of ambulance services alert the destination hospital within 5 minutes before reaching it. Air transport is used infrequently; 70% never use a helicopter.

Prehospital care and management

Seventy-six percent of services have a written protocol for prehospital stroke management (Table 2). The Cincinnati Stroke Scale (5) (or some modification of it) is the most commonly used stroke scale assessment tool. The stroke scale score is reported to hospital staff "sometimes" in 21% of cases, "always" in 34% of cases, and "rarely" or "never" in 36% of cases; 9% of respondents did not know whether the stroke scale score is reported to hospital staff.

Stroke training for EMS

A majority of respondents reported their personnel are trained on stroke at least once a year (119 [60%]) (Table 3). Nearly one-third (60 [30%]) reported their personnel receive stroke training once every 2 or 3 years. Classroom training was the most commonly cited format, followed by DVDs or videos (Table 3). Other types of training reported include run reviews and Advanced Cardiac Life Support refresher courses (6). Most respondents reported offering training opportunities annually (56%) or semiannually (24%) is most effective.

Forms of training reported most often as most effective include agency-sponsored conferences and seminars, regional conferences, Internet-based training, and statewide conferences (Figure). Agency-sponsored conferences or seminars was the kind of training most commonly reported effective for Emergency Medical Technician (EMT)-Basic (79%) and, to a lesser degree (32%), for paramedics. This form of training also was reported as effective for paramedics (32% of responses). Internet-based training was the second most common method for paramedics (22%), with regional (18%) and statewide (16%) conferences cited about equally. Teleconferences, videoconferences, and national conferences were least frequently cited as effective training vehicles.

Emergency departments

Eighty-four percent of the largest hospitals reported having written emergency department (ED) care protocols for acute stroke diagnosis or treatment, compared with 43% of small hospitals (Table 4). Fifty-two percent of rural hospitals have ED protocols. Of the hospitals with written stroke care protocols, 10 (17%) use telemedicine.

ED staff in approximately 80% of medium and large hospitals receive training for acute stroke diagnosis and treatment, compared with 60% of small hospitals and 65% of rural hospitals. Most of the hospitals with a stroke team are based in the Twin Cities metro area. One hundred twelve (93%) hospitals reported the ability to provide around-the-clock laboratory services, with expedited results within 45 minutes.

Among the rural hospitals, 16% do not have computerized tomography (CT) scan or a CT technician available around the clock (Table 4). Of the 17 hospitals that do not have access to a neurologist for consultation on stroke cases, all but one were rural hospitals; four hospitals reported having access to a neurologist for consultation during specific hours of the day.

Inpatient care

Approximately 27% of small, 58% of medium-sized, and 35% of rural hospitals have written inpatient protocols for managing stroke patients (Table 5). Fewer than 20% of small hospitals have a "stroke code" (alert system for hospital staff to respond to an inpatient stroke). Most (56) of the 60 hospitals that do not provide coordinated stroke care have fewer than 100 beds. Nearly every hospital that indicated not providing such coordinated stroke care reported being prepared to transfer stroke patients to an appropriate hospital for such care. Eight percent of rural hospitals, compared with 68% of Twin Cities metro hospitals, have a designated ward for stroke patients.

Quality improvement, education, and primary stroke center certification

At the time of the survey (January 2006), the Joint Commission had certified six Twin Cities metro hospitals as Primary Stroke Centers.

Forty-four (37%) hospitals reported their staff had sufficient opportunity to receive at least 8 hours each year of stroke education and training (Table 6). Staff in 10 (16%) of the 63 smallest hospitals have such an opportunity. About half of hospitals with 100 or more beds provide public education; 7% of small hospitals and 31% of medium-sized hospitals provide public education on stroke at least twice a year. Twenty percent of small hospitals and 42% of medium-sized hospitals have a database for tracking quality information. One-fourth of hospitals reported having a stroke champion working within the hospital as an advocate for improving stroke care.

Hospitals were asked to name their priorities for stroke, and written responses were grouped (Table 7). Development of protocols, standing orders, or pathways (31% overall) were most commonly reported, followed by early identification or assessment of stroke patients (20%) and staff education on stroke (19%). Other responses included rapid intervention or use of tissue plasminogen activator (tPA) (14%) and rapid transfer of stroke patients (13%) (Table 7).

Stroke hospitalizations and tPA questions

Seventy-two (61%) of 118 hospitals reported administering intravenous tPA for acute stroke treatment. Among the 74 small hospitals, 35 (49%) administer tPA, compared with 19 (73%) of 26 medium hospitals and 17 (89%) of 19 large hospitals. Among rural hospitals, 56 (56%) of 99 administer tPA, compared with 17 (89%) of 19 Twin Cities metro hospitals.

The MSP Steering Committee developed the following recommendations for consideration by ambulance services, hospitals, state agencies, and collaborating partner organizations in making systems-level improvements in stroke care. These suggestions and strategies were based on the results of these surveys and on discussion in quarterly meetings during 2006 after the surveys.

The results of these surveys will guide MDH and MSP in planning for systems-level intervention and programs to improve acute stroke care in Minnesota. In addition, ambulance services and hospitals can use these results to change their health systems to improve the quality of care for their stroke patients. Finally, these surveys highlight the need to develop a statewide or regional system for transportation and treatment of stroke patients to better facilitate coordination between EMS and hospital care. This system may involve development of formal or informal agreements to divert patients to appropriate facilities, establishment of a structure for providing continuing education to health care providers, and opportunities for health care organizations to improve their overall stroke care capacities.