A social marketing and communications company, Geovision, Inc, was hired to conduct secondary research on existing messages about stroke signs and symptoms. It conducted a comprehensive literature search and informal inventory of other unpublished sources. The results were used to develop an evidence-based approach to educating the lay public to recognize the signs of stroke and respond by calling 9-1-1. The search for messages and materials included online investigation and direct inquiry to national and international medical, public health, academic, and advocacy agencies and organizations that promote awareness, prevention, and treatment of stroke. These organizations included state and international health departments, hospitals and research centers, medical foundations, and universities. The search also included an assessment of barriers to calling 9-1-1 in response to acute medical events among the priority populations living in communities where the risk for stroke is elevated. Existing knowledge and strategies relating to the effective placement of health communication materials about medical emergencies were also assessed in the literature search.

Using results from the secondary research, Geovision created a 3-minute animation to teach the signs of stroke. We convened five groups to test the concept of the animation: three mixed-audience groups, one group of stroke educators, and one group of stroke survivors. All participants received financial compensation for attending a 90-minute session. We created three separate discussion guides to solicit comments on the animation concept and information content. Immediately after viewing the animation and before any discussion, the mixed groups completed a brief survey to evaluate information recall and message resonance.

The three mixed-audience groups consisted of 1) non-Hispanic white and non-Hispanic black men and women aged 63 or older who had not had a stroke; 2) non-Hispanic white and non-Hispanic black adult children of parents aged 65 or older; and 3) providers of community elder programs or non-health–care services. Because of high prevalence of stroke and stroke-related hospitalization and mortality in Fall River, Lowell, and Springfield, residents in these communities were recruited for the mixed-audience groups by professional telephone recruiters and community-based organizations. Participants provided information including age, household income levels, and employment status.

Twelve known stroke educators were contacted to participate in the stroke educators concept testing group. Additionally, education staff from over 40 hospitals and health care organizations were solicited to participate in the stroke educators group. Stroke survivors were recruited from a pool provided by MDPH, the Mt. Auburn Hospital Stroke Service Program, and the Watertown Senior Center.

Risk of stroke is higher for older and low-income women, and the rate of death from stroke more than triples with each added decade of life after the age of 45 (1). To assess message retention from our public education model, we conducted a pilot project with several cardiovascular disease risk reduction educators from the Well-Integrated Screening and Evaluation for Women Across the Nation (WISEWOMAN) program from the MDPH Women's Health Network. The network provides free breast and cervical cancer screening and diagnostic services to low-income women aged 40 to 64. Sites selected to participate in WISEWOMAN provide heart disease and stroke prevention services that include cardiovascular health risk assessment and lifestyle interventions by the cardiovascular risk reduction educators. The project proposal was submitted to and approved by the MDPH Human Research Review Committee. HSPC trained the educators, who in turn taught stroke signs and symptoms to a group of Women's Health Network clients.

Participants were asked to sign an informed consent form and provide basic demographic information. To assess changes in knowledge of stroke signs and symptoms, participants were given a pretest and posttest immediately before and after the education session. Participants received a $10 grocery store gift card for attending the session. Three months later, a follow-up telephone survey was conducted, after which participants received a $15 grocery store gift card.

Data were entered into Excel 2003 (Microsoft Corporation, Redmond, Washington) and then exported to SAS (SAS Institute, Inc, Cary, North Carolina) for analysis. Descriptive statistics were determined for demographic variables. Paired t-tests were used to determine differences in mean between pretest and posttest results as well as posttest and 3-month follow-up results. We considered the results of the paired t-tests statistically significant when the corresponding P value was ≤.05.

Geovision identified 24 private organizations and 20 national health departments and contacted them for sample materials. Three departments provided relevant program information and materials. Thirty articles were reviewed and summarized.

The search revealed that a comprehensive public awareness campaign that includes mass media can increase stroke recognition but should target family, coworkers, and caregivers of those at highest risk for stroke. Moreover, educational efforts should focus on behaviors that promote early seeking of hospital care.

The research, however, yielded a paucity of stroke education messages with evidence-based effectiveness. Among them was the Cincinnati Prehospital Stroke Scale (CPSS) (14), a three-item scale based on a simplification of the National Institutes of Health Stroke Scale (15), which has high sensitivity and specificity for identifying stroke patients who are candidates for thrombolytic treatment when performed by a physician (15) and has similar results when used by prehospital care providers (14). Moreover, the CPSS has been accurately administered by untrained laypeople to identify stroke signs in mock patients (16) and in stroke survivors (17) when prompted by a 9-1-1 telecommunicator. On further investigation, we learned that the CPSS had been successfully modified by adding a fourth item, so that it could be used by laypeople before they called 9-1-1. This item provided instructions for laypeople to call 9-1-1 if they observed stroke signs in someone. Kleindorfer et al created the FAST acronym for lay assessment of stroke symptoms (Table 1) (18). During a retrospective chart review of 3500 stroke patients, the authors found that the FAST message identified 88.9% of patients with stroke or transient ischemic attack (TIA) (i.e., brief stroke-like symptoms that can occur with no lasting damage) (18).

On the basis of its findings from the secondary research and on knowledge of basic memory techniques (e.g., mnemonics or acronyms, repetition, rhythm or songs), Geovision recommended creating a 3-minute musical animation using the FAST acronym. The animation was concept-tested and produced for English speakers.

Of 41 people recruited for the three mixed-audience groups, 34 participated. The mean age of participants was 58.1 years (Table 2), and they included seniors (41%), adults with senior parents (38%), and health care workers who care for seniors (21%). Twenty-two participants (65%) were female, and 23 (68%) were non-Hispanic whites. Forty-four percent of participants earned less than $20,000 per year.

Immediately after being shown the 3-minute animation, participants completed a brief form to assess information recall. From the three groups, 100% of participants indicated that the topic of the video was stroke and identified the action to take at the first sign of a stroke (call 9-1-1); 76.5% recalled the three symptom components of FAST (Face, Arm, and Speech).

Of the seven stroke survivors who participated in concept testing, three (43%) were female and all were non-Hispanic white (Table 3). The mean age was 54.1 years. Four (57%) had had ischemic strokes (i.e., strokes caused by a blocked blood vessel in the brain).

Nine stroke educators were part of the final concept-testing group. All nine were non-Hispanic white women, and they represented seven different hospitals or health care organizations. Their titles were diverse and included neuroscience educator and director of rehabilitation services.

Participants from all five groups thought the animation provided a comfortable, upbeat, "less scary" approach to a serious health issue, and that the animation and music were simple, direct, and uncluttered. Participants consistently described the accompanying song as "catchy" and the animation style as simple but appealing. Recommendations included creating a visual reinforcement of the FAST checklist, depicting a younger character in at least one of the scenarios, and accentuating the distorted speech bubble. Some stroke educators expressed concern about showing only the three most common symptoms without suggesting that other symptoms may indicate stroke.

We incorporated suggestions from the concept testing into a final animation and created a brochure and poster (Figure) using the characters from the animation. These pieces were included in a Stroke Heroes Act FAST educational kit, which had instructive PowerPoint presentations with information about stroke signs, symptoms, and risk factors; an educator's guide with strategies for effectively reaching lay audiences; a press release; and evaluation forms. The kit, which we designed as a train-the-trainer module, is available through HSPC.

Seventy-two women participated in the retention pilot. One participant refused to provide demographic information. The mean age of participants was 53.7. Participants were primarily non-Hispanic whites (76%) and had never had a stroke (99%). Almost half (49%) of participants had a family history of stroke (Table 4).

Immediately after the education session, we saw significant increases in the percentage of participants who recognized that facial droop (91.7% vs 98.6%, P = .02) and arm weakness or numbness (86.1% vs 97.2%, P = .004) were symptoms of stroke. Moreover, we also saw significant increases in the percentage of participants who thought chest pain and shortness of breath were not stroke symptoms. The percentage of participants who recognized "call 9-1-1" as the action to take after observing stroke symptoms also increased significantly, from 81.9% to 98.6% (P < .001) (Table 5).

Three months after the education session, we conducted a follow-up survey of the participants. Seven participants could not be reached. Of the 65 participants who were given the survey, 100% remembered slurred speech and facial drooping as symptoms of stroke; 98.5% remembered arm weakness or numbness; 96.9% would call 9-1-1 if they thought someone was having a stroke. None of these is a significant change from the posttest (Table 6). However, significantly fewer participants remembered correctly that chest pain (87.7% vs 55.4%, P < .001) and shortness of breath (84.6% vs 50.8%, P < .001) were not symptoms of stroke. Moreover, the percentage of participants who recognized all components of the FAST acronym declined significantly, from 100.0% to 78.5% (P < .001).