Public health efforts to promote stroke awareness and the need to seek urgent treatment have assumed a new importance in the years since the publication of a major clinical trial showing decreased short-term disability and improved outcomes for patients experiencing an ischemic stroke after thrombolytic therapy (1). Prehospital barriers to prompt treatment for ischemic stroke include the lack of awareness of stroke warning signs in patients and family members, underuse of 911 emergency medical services (EMS), and long distances to tertiary-care facilities that provide diagnostic and treatment services (2-5). These barriers can lead to delayed presentation to the emergency department and to ineligibility for time-dependent treatment. Achieving increased use of thrombolytic therapy within the three-hour window in one community required a multilevel intervention to influence the knowledge and behavior of the public, the response of EMS, and the coordination of diagnostic and treatment facilities at the hospitals (6,7).

From a public health perspective, an important component for the success of stroke interventions is to improve public knowledge about stroke, particularly focusing on individuals at high risk and their family members and caregivers. Public education campaigns in selected communities have been effective in increasing the level of knowledge about stroke, but most data have come from large metropolitan centers working with academic institutions (7,8). Much less is known about basic knowledge of stroke symptoms, risk factors, and the need for urgent intervention among residents in rural communities that are relatively isolated from major metropolitan centers.

We conducted a telephone survey in two rural counties in Montana in 2004. This report describes the level of awareness of stroke warning signs and risk factors and the public perception of the need to call 911 EMS for stroke in residents aged 45 years and older.

The mean age of respondents (N = 800) was 61 years (range 45 to 95); 60% were female; 96% were white; 2% were American Indian; and 93% reported 12 or more years of education. Ten percent reported a history of atrial fibrillation; 6% reported a history of diabetes; 37% reported a history of high blood pressure; 31% reported a history of high cholesterol; 17% currently smoked cigarettes; and 40% were former smokers. Eight percent reported a history of heart disease, and 6% reported a history of stroke or TIA. Overall, 80% reported one or more risk factors for stroke, and 56% reported two or more risk factors for stroke.

Numbness on any side of the face or body (45%) and speech problems (38%) were the most frequently reported established warning signs for stroke (Table 1). Fewer respondents reported vision problems (18%) or difficulty walking (11%). Overweight (56%), smoking (50%), and high blood pressure (44%) were the most frequently reported risk factors for stroke (Table 1).

The majority of respondents (70%) could identify two or more warning signs for stroke (Table 2). Women (75%) were more likely than men (62%) to identify two or more established warning signs for stroke, and respondents aged 45 to 64 years (76%) were more likely than those aged 65 years and older (59%) to identify two or more established warning signs for stroke. Just under half of the respondents (45%) could identify two or more established risk factors for stroke. Respondents aged 45 to 64 years (48%) were more likely to identify two or more established risk factors for stroke compared with those aged 65 years and older (40%).

Adjusting for multiple factors using logistic regression analyses, women, individuals aged 45 to 64 years, those with 12 or more years of education, and individuals with a history of high cholesterol were more likely to identify two or more established warning signs for stroke compared with respondents without these characteristics (Table 3). Women and respondents aged 45 to 64 years were more likely to identify two or more established risk factors for stroke compared with men and with respondents aged 65 years and older.

Overall, the majority of respondents (76%) indicated they would call 911 EMS if they witnessed someone having a stroke (Table 4). There were no differences by age, sex, or years of education in the proportion of respondents who indicated they would call 911 if they witnessed someone having a stroke (data not shown). When asked what they would do if they were experiencing sudden difficulty speaking, numbness, or weakness or paralysis, 43% to 49% of individuals indicated they would call 911. Depending on the symptom, 17% to 23% indicated they would go to the hospital, 14% to 19% would call their doctor, 11% to 18% would call their spouse or a family member, and 3% to 5% would do something else (Table 4). Respondents aged 65 years and older were more likely than respondents aged less than 65 to indicate they would call 911 if they experienced sudden difficulty speaking (49% vs 42%, P = .04), numbness (50% vs 40%, P = .006), or weakness or paralysis (55% vs 45%, P = .004). There were no differences by sex or years of education in the proportion of respondents who indicated they would call 911 if they experienced any of these warning signs (data not shown).

The majority of respondents from these rural counties were aware of the established warning signs for stroke, and awareness was higher in women, younger respondents, those with a higher level of education, and those with a history of high cholesterol compared with respondents without these characteristics. Interestingly, respondents with a history of other major stroke risk factors (e.g., high blood pressure) were no more aware of the warning signs compared with respondents without these conditions. Overall, fewer respondents were aware of the established risk factors for stroke. We also found that the majority of respondents would call 911 if they thought someone was having a stroke, but less than half would call 911 if they were experiencing stroke warning signs.

Individuals responding to the survey lived in communities that are typical of many communities across the United States, where health care for a large region is centered in a nearby town. Awareness of established stroke warning signs and risk factors was higher than awareness levels reported in other community surveys (10,16,17). In 1999, only 30% of adults surveyed in Michigan identified two or more warning signs correctly, and 34% identified two or more risk factors correctly (17). In 2000, more than 40% of respondents in Cincinnati, Ohio, identified two or more warning signs for stroke, and 32% identified two or more established risk factors for stroke (16). A study of adults living in an urban area of Australia in 1999 found that 26% of respondents to a telephone survey identified two or more warning signs for stroke, and 50% identified two or more risk factors for stroke (10).

Our findings on what respondents would do if they witnessed someone having a stroke or if they were experiencing warning signs of stroke are comparable to previous studies from Australia and Michigan (10,17). In Australia, 67% of respondents would call an ambulance if they witnessed someone having a stroke, while less than half of respondents would call an ambulance if they were experiencing sudden stroke warning signs (45% difficulties with speech, 38% numbness/weakness, 35% weakness or paralysis). In Michigan, however, 79% indicated they would call 911 if someone was having a stroke.

There are a number of limitations to this study. First, the survey does not reflect the experience of residents without telephones. Second, self-reported information regarding risk factors for stroke is subject to recall bias. Previous studies, however, have found that self-reported risk factors for cardiovascular disease are reported reliably (18,19). Third, respondents were asked unaided questions to assess respondent knowledge of the warning signs and risk factors for stroke. Some previous studies assessing awareness of stroke warning signs used aided questions and found higher levels of knowledge than the levels found in this study (20). It is possible that unaided questions may underestimate the awareness of the warning signs of stroke, and aided questions may overestimate awareness. Fourth, this study was conducted in a rural non-Hispanic white population, and there may be significant variation in awareness of stroke warning signs and risk factors in other geographic and racial and ethnic communities in the United States.   

Our findings about stroke awareness in rural communities are important because they are similar to studies published from academic centers working in urban areas (10,16,17). This effort, however, represents a collaboration between a state public health agency and several regional health care systems not only to understand the levels of stroke awareness in two communities at baseline but also to promote community awareness and increased use of EMS and to define regional approaches for prompt stroke treatment in cooperation with the stroke referral centers in these counties. Based on the findings reported here, the Montana Cardiovascular Health Program and these partners have developed and implemented a multifaceted intervention to increase community awareness of the warning signs of stroke and the need to use 911 adapting strategies that have been shown to be successful (8,21).

It is reassuring that levels of community awareness and knowledge about stroke in rural settings are not markedly different from levels found in urban environments. Others have shown that multifaceted interventions to increase community awareness, use of EMS, and availability of prompt diagnosis and treatment can succeed (7,8). Strategies to reduce barriers to prompt stroke treatment may be somewhat different across large frontier areas compared with urban environments. Cooperative efforts between public health agencies, communities, and prehospital and acute care systems, however, can build on the published experience of others to understand and improve knowledge about stroke warning signs and risk factors as part of broader public health interventions targeted to stroke prevention and treatment. In the United States, public health programs are only beginning to develop partnerships and work toward developing and documenting successful interventions in a wide variety of settings. This report represents the beginning of efforts to meet the challenges in rural communities in Montana and across many rural areas.