Thrombolytic therapy is an effective but time-sensitive treatment of acute ischemic stroke. Delay in arriving at the emergency department (prehospital delay) is an important reason thrombolysis is underused.1 African Americans have higher rates of stroke,2 poststroke disability,3 and prehospital delay1 than individuals of other races and ethnicities.
Seeking help for acute ischemic stroke symptoms is a complex process. One factor that may be amenable to influence is recognition that symptoms are due to stroke, which predicts earlier hospital arrival.4 One-fifth of African American women are unable to name a single stroke symptom,5 and they may perceive symptoms as similar to other health problems or everyday bodily sensations.6 Stroke preparedness interventions aim to increase awareness of stroke symptoms and behavioral intent to contact emergency medical services (EMS), but few interventions target African American women. Training beauticians to educate clients improved stroke awareness among African American women,7 and peer-led workshops were effective for African Americans of both sexes,8 but other interventions showed inconsistent results.9
Computer-based health education shows promise for cardiovascular knowledge acquisition.10 In a Pew Center survey, 72% of respondents sought health information online with women more likely to investigate the cause of specific symptoms.11 Most of African Americans (78%) use the Internet.12 However, more than half of consumer-oriented stroke Web sites are written higher than the recommended grade 6 reading level.13 The purpose of this pilot study was to evaluate the effect of Sisters Against Stroke, an Internet-based stroke preparedness intervention for African American women, for (1) self-efficacy to recognize a stroke and to know what to do in the event of stroke, (2) awareness of stroke symptoms, and (3) identification of appropriate action in response to stroke symptoms.
Methods
Intervention
Sisters Against Stroke presents information about symptoms of acute ischemic stroke and action to take in the event of stroke and incorporates key elements for stroke preparedness: symptom recognition, importance of immediate action (call 911), and availability of treatment. Self-efficacy theory14 as seen through the lens of womanist ways of knowing epistemology15 guided intervention development.
Performance accomplishment, vicarious experience, and verbal persuasion are constructs theorized to increase self-efficacy.14 Presenting thrombolytic treatment in the context of benefits to women and their families of immediately calling 911 for suspected stroke is an example of verbal persuasion. Vicarious experience was accomplished with videos of African American actors depicting stroke onset in familiar settings such as a supermarket. The videos and lay language descriptions of symptoms emphasize the effect of stroke on daily activities, which is consistent with a womanist epistemology to incorporate into health interventions everyday experiences in the lives of African American women.15
A 6-member community advisory group consisting of the director of the city/county public health district, first ladies of 2 churches with predominantly African America membership, and women representing African American community organizations guided intervention development and pilot testing. The group first evaluated findings from a formative qualitative study to ensure the intervention addressed identified stroke education needs, particularly experiential depictions of stroke onset,6 and that it was culturally sensitive. At subsequent meetings, the group discussed participant inclusion criteria and recruitment strategies and identified churches for the pilot study. Group members elicited support from pastors and women in positions of church leadership and facilitated recruitment at the churches.
Sisters Against Stroke has a custom logo and layout with images of African American women. The site features American Heart Association/American Stroke Association (AHA/ASA) “Spot a Stroke FAST symptoms” (face drooping, arm weakness, speech difficulty), which capture 88.9% of symptoms.16 The intervention conforms to national guidelines for online consumer health information. A clinical nurse specialist and a stroke neurologist with expertise, respectively, in patient education and stroke preparedness interventions for African Americans reviewed the site during development and provided suggestions for changes.
Design and Sample
The study received exempt status from the institutional review board. The nonprobability sample consisted of African American women aged 35 and older who use computer and email. Recruitment took place at 4 churches. The investigator sent an email to 66 women who indicated interest, 48 of whom responded. Of the 48 respondents, 44 agreed to participate. A quasi-experimental design was used with the opaque sealed envelope method to assign 2 churches to the intervention group and 2 churches to the wait-list comparison group. Randomization at the level of the church reduced possibility of diffusion of treatment.
At pretest, 47.7% of the participants indicated they would not be able to tell if someone was having a stroke.
Informed consent and data collection were online. Demographic characteristics, perception of stroke preparedness, and sources of stroke information were collected at the start of the study (pretest). All other measures were collected at the pretest, immediately after intervention group access to Sisters Against Stroke (posttest), and 1 month later. Both research groups completed measures on the same schedule. The intervention group had access to Sisters Against Stroke for 2 weeks after the pretest. Comparison group participants had access to the intervention for 2 weeks after completing the 1-month posttest. Participants received a $25 gift card at each data collection point.
Measures
Awareness of Stroke Signs and Symptoms
Ability to name stroke symptoms was measured with an open-ended question from an AHA/ASA national survey: “Based on what you know, what warning signs do you associate with having a stroke?”17 One point was allotted for each Spot a Stroke FAST symptom (face drooping, arm weakness, speech difficulty) named. Scores range from 0 (no FAST symptoms named) to 3 (3 FAST symptoms named).
Ability to recognize AHA/ASA Spot a Stroke FAST symptoms also was measured with an instrument consisting of brief written vignettes describing symptoms in the context of everyday activities and in familiar settings (eg, church).18 The original version of the instrument has 12 stroke vignettes (hemiparesis, dysarthria, aphasia, visual changes, dizziness/imbalance, and severe headache) and 4 nonstroke vignettes (chest pain, musculoskeletal pain, orthostatic hypotension, and epistaxis). Internal consistency reliability for stroke recognition was 0.88.18 For the current study, the 8 vignettes depicting Stop a Stroke FAST symptoms and 4 nonstroke vignettes were used. Participants selected “stroke,” “no stroke,” or “do not know” for each vignette. One point is allotted for each correctly identified vignette. Scores range from 0 (no stroke vignettes correctly identified) to 8 (all stroke vignettes correctly identified). Cronbach’s α for the current sample was .78.
Self-efficacy to Recognize Stroke Symptoms and Take Correct Action
Self-efficacy was measured with 2 statements from a study of stroke awareness among African Americans: “I would not be able to tell if someone is having a stroke” and “If I saw someone having a stroke I would not know what to do.”18 Responses were a 4-point Likert-type scale ranging from “strongly agree” to “strongly disagree.”
Appropriate Action in Response to Stroke Symptoms
Two open-ended questions from an AHA/ASA survey assessed behavioral intent to call 911 for suspected stroke: “If you thought someone was experiencing signs of a stroke what is the first thing you would do?” and “If you thought you were experiencing signs of a stroke what is the first thing you would do?”17 Frequency of “call 911” responses was calculated for each question.
Behavioral intent to call 911 also was measured with the stroke vignette instrument. Each vignette concludes with “If this happened, what would you do first?” followed by 4 options (call doctor’s office immediately, wait a couple of hours and then decide, call a family member or friend immediately, or call 911 immediately).18 Each “call 911 immediately” response is 1 point. Scores range from 0 (no “call 911 immediately” responses) to 8 (“call 911 immediately” chosen for all 8 stroke vignettes). Internal consistency reliability of the original instrument was 0.83.18 Cronbach’s α for the current sample is .70.
Demographic Characteristics and Perception of Stroke Preparedness
Participants provided demographic characteristics (age, education, employment, marital status) and stroke risk factors (high blood pressure, high cholesterol, heart disease, diabetes, 20 lb over ideal weight, physical inactivity, cigarette smoking). Perception of stroke preparedness was assessed with an AHA/ASA survey item: “How informed are you about stroke or ‘brain attack’ in women?”17 Responses were a 4-point Likert-type scale ranging from “very well informed” to “not at all informed.” Sources of stroke information were selected from a 9-item list (brochure, billboard, TV program or commercial, radio, Internet, book, health fair, doctor, or nurse).
Statistical Analysis
Data were entered into SAS version 9.4. Descriptive statistics including frequency distributions, percentages, means, and standard deviations were calculated for participant characteristics and study variables. Between-group differences for demographic characteristics, stroke risk factors, sources of stroke information, and extent to which participants felt informed about stroke were analyzed using the t test for independent groups for interval level data and Cochran-Mantel-Haenszel (CMH) statistics for nominal and ordinal level data. Cochran-Mantel-Haenszel statistics do not require distributional assumptions and provide stratified analysis of the relationship between row and column variables while controlling for variables in a multiway table.19 Within-group differences for ordinal level dependent variables and interval level variables that did not meet normality assumptions were analyzed using Friedman test and post hoc analysis with Wilcoxon signed rank test. Cochran-Mantel-Haenszel statistics were used to analyze nominal level dependent variables. Effect size estimates were calculated by converting z scores to r. Statistical significance was set at α = .05. Intent-to-treat analysis was used. Missing values for 3 participants who did not fill out time 3 measures were replaced using “last observation carried forward” approach.
Results
Sample Characteristics
The mean (SD) age of the 44 participants was 52.98 (9.26) (range, 35–70) years. Most were married (52.27%) and employed full time (74.42%). The comparison group was older (t = 2.16, P = .03) with a higher education level (χ2CMH = 6.07, P = .01). Stroke risk factors reported most frequently were overweight (70.45%) and hypertension (54.55%). Women in the comparison group were more likely to report hypertension (χ2CMH = 4.51, P = .03) (Supplemental Digital Content 1, available at http://links.lww.com/JNN/A90).
One-fifth of the sample (20.45%) considered themselves not at all informed about stroke, and 65.91% considered themselves moderately informed. The most frequent sources of stroke information were Internet (63.64%), brochures (60.47%), health fairs (52.27%), and health providers (52.27%).
Research Question 1: Self-efficacy to Recognize a Stroke and Know What to Do in the Event of Stroke
At pretest, 21 of the 44 participants (47.7%) indicated that they would not be able to tell whether someone was having a stroke (Table 1), with no significant improvement in self-efficacy to recognize stroke. More than half of the sample (56.8%) indicated that they would not know what to do in response to suspected stroke symptoms. The intervention group showed improved self-efficacy to know what to do in the event of stroke at posttest (Z = −3.090, P = .002, r = −0.46) and 1 month later (Z = −3.34, P = .001, r = −0.49). The comparison group showed improved self-efficacy to know what to do in the event of stroke, perhaps indicting testing effect.
Research Question 2: Awareness of Stroke Symptoms
The mean number of AHA/ASA Stop a Stroke FAST symptoms named by the sample in response to an open-ended question at pretest was 1.65 (Table 2), with no significant improvement over time. The mean number of stroke symptom vignettes correctly identified was 5.5 of 8 stroke vignettes. The intervention group showed improvement in ability to identify stroke symptoms in the vignette measure at posttest (ΖZ = −2.95, P = .003, r = −0.43) and 1 month later (ΖZ = −3.381, P = .001, r = −0.50).
Research Question 3: Correct Action in Response to Stroke Symptoms
When asked in an open-ended question what participants would do if they thought they were having a stroke, half of the sample (50%) reported at pretest that they would call 911 (Table 2). The intervention group increased in the number of women indicating that they would call 911 for suspected stroke symptoms in themselves at posttest (χ2CMH = 5.71, P = .01) and 1 month later (χ2CMH = 5.71, P = .01). The comparison group increased in calling 911 for themselves at the 1-month posttest, perhaps indicating testing effect.
In response to an open-ended question asking what participants would do if they thought someone other than themselves might be having a stroke, the majority of the sample (81.8%) stated at pretest that they would call 911 (Table 2). The intervention group increased at posttest in the number of women indicating that they would call 911 for suspected stroke in someone else (χ2CMH = 5.48, P = .01).
With regard to identifying correct action in response to stroke symptom vignettes, at pretest, the mean number of correct responses (call 911) to the stroke symptom vignette instrument was 5.8 of 8 vignettes depicting stroke (Table 2). The intervention group showed significant improvement in identifying correct action (call 911) in response to stroke symptom vignettes at posttest (ΖZ = −2.179, P = .02, r = −0.32) and 1 month later (ΖZ = −3.018, P = .003, r = −0.44).
Discussion
This pilot study showed preliminary effectiveness of an Internet stroke preparedness intervention for African American women that presented stroke symptoms experientially through written description and videos. Small to medium effect sizes were obtained. A limitation is a volunteer sample and possible selection bias. A history effect is possible due to Spot a Stroke FAST billboards coinciding with the study.
The intervention used lay language to describe symptoms in addition to Spot a Stroke FAST terminology. Lack of improvement naming symptoms may reflect difficulty recalling information in response to open-ended questions. Significant intervention group effect for recognizing symptoms on an instrument that describes stroke symptoms in vignettes seems aligned with previous research in which individuals were more adept recognizing stroke symptoms described in lay language than in terminology from public educational materials.20 Future studies comparing formats to deliver information about stroke symptoms could help determine the most effective approach for African American women.
Behavioral intent to call 911 for someone else was high when the diagnosis of stroke was given in an open-ended question. However, previous research showed significant disparity between intent to call 911 and actual EMS use.21 Identification of correct action in response to lay language descriptions of symptoms may be a better indicator of behavioral intent than an open-ended question with a given diagnosis of stroke because the former requires individuals to connect a judgment about the cause of a specific symptom with an action.20
Barriers to calling 911 previously identified by African Americans include cost and variability in EMS response time.22 Stroke preparedness interventions can counter perceptions that private vehicle transport results in quicker hospital arrival with information that persons arriving by ambulance are seen sooner at the emergency department.23 Information about prioritization of 911 calls for symptoms suggestive of stroke may further address perceptions about EMS response times, but concerns about cost are unlikely to be alleviated until more communities subsidize EMS transport.22
Of concern is the large number of participants at baseline who would not call 911 for themselves but would do so for someone else. A possible explanation is gender roles that prioritize women taking care of other people,24 rendering barriers to calling 911 less salient when another person is having symptoms. Responses about calling 911 for oneself also may reflect beliefs that persons with stroke are unable to call 911.25 A significant intervention group increase in intent to call 911 for themselves may reflect intervention content that addresses hesitations, such as embarrassment, about activating EMS for oneself. Because the comparison group also improved in intent to call 911 for oneself, testing effect was possible.
The Internet was the most frequent source of stroke information for this sample, with a mean age of 52 years. Lack of computer access or proficiency limits the reach of Internet stroke preparedness interventions, particularly for older individuals who are less likely to use the Internet. Pairing computer-proficient women with women less comfortable with technology may increase participation of older women. Anecdotal evidence that participants viewed the study Web site and completed measures on smart phones suggests another avenue to expand intervention reach.
This pilot study is one of very few stroke preparedness intervention targeting African American women. Nurses are uniquely qualified to promote Healthy People 2020 goals to improve stroke awareness due to their contact with at-risk individuals in a wide variety of inpatient and outpatient settings and nursing’s disciplinary emphasis on health education. Internet-based stroke preparedness interventions that emphasize experiential descriptions of stroke onset may be a feasible and cost-effective addition to public stroke preparedness efforts and a way to supplement face-to-face education when time constraints preclude stroke education during patient encounters.
Acknowledgments
The author gratefully acknowledges community advisory committee members Sheila Blanton, Candance Hayes, Pamelia Hytche-Hunter, Mia Moody-Ramirez, Connie Nichols, and Sherry Williams and the women who participated in the study. Gratitude to Lesli Skolarus, MD, Department of Neurology, University of Michigan, for sharing the stroke symptom videos and to Jack Tubbs, PhD, and Johanie Van Zyl, MS, Baylor University Department of Statistics, for assistance with statistical analysis.
References
1. Tong D, Reeves MJ, Hernandez AF, et al. Times from symptom onset to hospital arrival in the get with the Guidelines–Stroke program 2002 to 2009: temporal trends and implications.
Stroke. 2012;43(7):1912–1917.
2. Howard VJ, Kleindorfer DO, Judd SE, et al. Disparities in stroke incidence contributing to disparities in stroke mortality.
Ann Neurol. 2011;69(4):619–627.
3. Burke JF, Freedman VA, Lisabeth LD, Brown DL, Haggins A, Skolarus LE. Racial differences in disability after stroke: results from a nationwide study.
Neurology. 2014;83(5):390–397.
4. Mandelzweig L, Goldbourt U, Boyko V, Tanne D. Perceptual, social, and behavioral factors associated with delays in seeking medical care in patients with symptoms of acute stroke.
Stroke. 2006;37(5):1248–1253.
5. Mochari-Greenberger H, Towfighi A, Mosca L. National women’s knowledge of stroke warning signs, overall and by race/ethnic group.
Stroke. 2014;45(4):1180–1182.
6. Beal CC. Stroke education needs of African American women.
Public Health Nurs. 2015;32(1):24–33.
7. Kleindorfer D, Miller R, Sailor-Smith S, Moomaw CJ, Khoury J, Frankel M. The challenges of community-based research: the beauty shop stroke education project.
Stroke. 2008;39(8):2331–2335.
9. Boden-Albala B, Stillman J, Roberts ET, et al. Comparison of acute stroke preparedness strategies to decrease emergency department arrival time in a multiethnic cohort: the stroke warning information and faster treatment study.
Stroke. 2015;46(7):1806–1812.
10. Beranova E, Sykes C. A systematic review of computer-based softwares for educating patients with coronary heart disease.
Patient Educ Couns. 2007;66(1):21–28.
13. Sharma N, Tridimas A, Fitzsimmons PR. A readability assessment of online stroke information.
J Stroke Cerebrovasc Dis. 2014;23(6):1362–1367.
14. Bandura A. Self-efficacy: toward a unifying theory of behavior change.
Psychol Rev. 1977;84(2):191–215.
15. Banks-Wallace J. Womanist ways of knowing: theoretical considerations for research with African American women.
Adv Nurs Sci. 2000;22(3):33–45.
16. Kleindorfer DO, Miller R, Moomaw CJ, et al. Designing a message for public education regarding stroke: does FAST capture enough stroke?
Stroke. 2007;38(10):2864–2868.
17. Mosca L, Hammond G, Mochari-Greenberger H, Towfighi A, Albert MA; American Heart Association Cardiovascular Disease and Stroke in Women and Special Populations Committee of the Council on Clinical Cardiology, Council on Epidemiology and Prevention, Council on Cardiovascular Nursing, Council on High Blood. Fifteen-year trends in awareness of heart disease in women: results of a 2012 American Heart Association national survey.
Circulation. 2013;127(11):1254–1263.
18. Skolarus LE, Zimmerman MA, Murphy J, et al. Community-based participatory research: a new approach to engaging community member to rapidly call 911 for stroke.
Stroke. 2011;42(7):1862–1866.
19. Agresti A. C
ategorical Data Analysis. 3rd ed. Hoboken, NJ: John Wiley & Sons, Inc; 2013.
20. Billings-Gagliardi S, Mazor KM. Development and validation of the stroke action test.
Stroke. 2005;36(5):1035–1039.
21. Hsia AW, Castle A, Wing JJ, et al. Understanding reasons for delay in seeking acute stroke care in an underserved urban population.
Stroke. 2011;42(6):1697–1701.
22. Skolarus LE, Murphy JB, Zimmerman MA, et al. Individual and community determinants of calling 911 for stroke among African Americans in an urban community.
Circ Cardiovasc Qual Outcomes. 2013;6(3);278–283.
23. Morris DL, Rosamond W, Madden K, Schultz C, Hamilton S. Prehospital and emergency department delays after acute stroke: the Genentech stroke presentation survey.
Stroke. 2000;31(11):2585–2590.
24. Woods-Giscombé CL. Superwoman schema: African American women’s views on stress, strength, and health.
Qual Health Res. 2010;20(5):668–683.
25. Beal CC, Stuifbergen A, Volker D. A narrative study of women’s early symptom experience of ischemic stroke.
J Cardiovasc Nurs. 2012;27(3):240–252.
