Kitko, Lisa; Hupcey, Judith E.
Stroke is the third leading cause of death and the leading cause of serious, long‐term disability in the United States. In 2004, there were more than 150,000 stroke‐related deaths, accounting for 1 of every 16 deaths in the United States (American Heart Association [AHA], 2007). About 700,000 people in the United States sustain a new or recurrent stroke each year, with an estimated 2007 cost of $62.7 billion (AHA).
The most significant cost of stroke is the human cost. Up to 50% of strokes result in death; those who do survive often are disabled (AHA, 2007). According to the National Heart, Lung, and Blood Institute, 6 months after an ischemic stroke, 30% of individuals 65 years of age or older are unable to walk without assistance, 26% are dependent in activities of daily living, and 26% are in nursing homes (Kelly‐Hayes et al., 2003). Although stroke is devastating and its impact is widespread, many people have limited knowledge about stroke symptoms and available treatments.
Historically, no treatment was available for ischemic stroke patients other than stabilization and attempts to prevent secondary complications. This led to a “wait and watch” approach; however, with recent advances in acute‐stroke treatment, this approach is being abandoned. Recombinant tissue plasminogen activator (rt‐PA) was approved in 1996 for the treatment of acute stroke. It significantly reduced mortality and disability, but the drug has to be administered within 3 hours of symptom onset to be effective (National Institute of Neurological Disorders and Stroke rt‐PA Stroke Study Group [NINDS rt‐PA Stroke Study Group], 1995). Thrombolytic efficacy is reduced with time. The longer the delay between the onset of symptoms and treatment with thrombolytics, the lower the chance of restoring blood flow and the higher the risk of hemorrhagic complications (Hacke et al., 2004).
Delay to presentation from the onset of symptoms to arrival at the emergency department (ED) continues to be the primary cause of patient exclusion from interventional thrombolytic stroke trials (Spilker, 1996). Registry data indicate that fewer than 2% of patients with acute ischemic stroke currently receive thrombolytic therapy (Katzan et al., 2000). Efforts to reduce therapy‐delay time for acute stroke may be more effective if the factors that delay hospital‐arrival time are identified.
Researchers have documented that fewer than 50% of stroke patients seek care within 3 hours.
Few studies have extensively examined the factors associated with stroke treatment delays in the United States. Researchers have documented that fewer than 50% of stroke patients seek care within 3 hours. Kothari and colleagues (1999) found only 30% of stroke patients are admitted within 3 hours, and Williams, Bruno, Rouch, and Marriott (1997) found that fewer than 25% arrived within 3 hours. These studies consistently have concluded that most stroke patients are unaware of the warning signs of stroke and present late because they misjudge the seriousness of their symptoms. Even when patients know they are having a stroke, most do not seek immediate medical attention (Williams et al.).
The presentation time between symptom onset and arrival at a care site is a critical factor in determining the treatments available for patients who sustain acute stroke. Delays in presentation can dramatically reduce the number of available treatment options. Minimal research exists on the factors that influence a person's decision to seek treatment.
In a study by Alberts, Bertels, and Dawson (1990), a patient's lack of knowledge about stroke symptoms was identified as a reason for delays in seeking treatment. According to the researchers, only 42% of patients presented within 24 hours of stroke onset, while 25% presented within 48 hours and 33% presented after 48 hours. It is critical to determine what factors influence delay time so effective interventions can be developed to reduce patient response times and potentially improve outcomes.
The purpose of this study was to describe factors influencing the decision‐making process of seeking treatment for acute‐stroke symptoms. If specific factors that delay hospital arrival time can be understood from the patient's perspective, these factors can be targeted for specific interventions to increase the number of patients who seek treatment within the therapeutic time window.
The Health Belief Model (HBM), as developed by Hochbaum, Leventhal, Kegeles, and Rosenstock (Becker, 1974), is used as the theoretical underpinning for understanding health‐seeking behaviors. The HBM is the dominant paradigm for describing determinants of preventive or protective behavior. The HBM evolved from the premise that each person's perception of the world determines what that person will do. The concepts of the model include a person's perceptions of susceptibility to the disease, the severity or seriousness of the disease, and the benefits and barriers associated with the choice of action that may prevent the disease process. According to Maiman and Becker (1974), the HBM hypothesizes that health‐related action depends upon the simultaneous occurrence of three classes of factors: the existence of sufficient motivation (or health concern) to make health issues salient or relevant; the belief that one is susceptible (vulnerable) to a serious health problem or to the sequelae of that illness or condition (perceived threat); and the belief that following a particular health recommendation would be beneficial in reducing the perceived threat at a subjectively acceptable cost. Cost refers to perceived barriers that must be overcome to follow a health recommendation. The model is disease‐ and action‐specific, focusing on behavior that decreases the probability of encountering illness. The theory was developed to explain preventive health behaviors.
The HBM provides a framework for determining the predictive variables that influence healthcare choices for acute‐stroke patients or their family members when they are faced with the signs and symptoms of acute stroke.
This qualitative study explored factors involved in the decision‐making process of seeking treatment when a person had symptoms of an acute ischemic stroke. More specifically:
* Who, if anyone, helped influence the patient's decision to seek care?
* How did the initial symptoms influence the patient's decision to seek care?
* What other factors influenced the patient's seeking care?
The study was undertaken using the methods of grounded theory, in which data collection and analysis occur simultaneously (Glaser & Strauss, 1967). The primary purpose of grounded theory is to generate explanatory models of human behavior that are grounded in the data (Morse & Field, 1995). Qualitative methodology was selected because it provides the flexibility and freedom to explore life experiences in an in‐depth manner.
Patients with a diagnosis of acute ischemic stroke regardless of the time from stroke to admission were recruited from a community hospital. Inclusion criteria included having sustained an acute cortical ischemic stroke with a National Institutes of Health Stroke Scale (NIHSS) score of 5‐22, or being the caregiver who was involved in the decision‐making process to seek care if the patient was unable to participate in the interview process due to medical complications such as aphasia or clinical presentation.
Sixteen patients met the inclusion criteria and were approached, but only 10 consented to participate. Reasons given for not participating included not wanting to talk into a tape recorder, denial of stroke, getting ready for a transfer, fatigue, or embarrassment as a result of slurred speech.
Before beginning the study, approval was obtained from both the university's and the hospital's institutional review board. The investigator obtained written informed consent before initiating data collection.
All data were kept strictly confidential. Interviews were transcribed verbatim by a trained transcriptionist who signed a letter of confidentiality. The participants were identified by a letter of the alphabet assigned by the researcher at the beginning of the interview. Audiotapes and consents were kept in a secure, locked area.
The interviews were conducted in the patient's room with either the patient alone or with both the patient and caregiver present. In several interviews, the patient served to validate the information provided by the caregiver. In one case, the patient had expressive aphasia, but would nod her head in response to her husband's answers. The taped interviews lasted 20‐30 minutes. The unstructured interviews included open‐ended questions to offer the researcher flexibility in gathering information. The interview began with demographic questions about gender, age, race, living arrangements, being alone at the time of stroke, distance from a hospital, NIHSS score, type of stroke, time from symptoms to arrival in the ED, method of transportation to the hospital, history of stroke, and family history of stroke. Cognitive responses were validated during this phase, during which patients could correctly identify the demographic information of date of birth, address, and home phone number. The interview then proceeded with the question “Why did you come to the hospital?” Participants were allowed to answer in a narrative fashion. An interview guide with probing questions was used to gather more detailed information. These questions included:
* When did you first experience symptoms and what were they?
* What did you think might be the cause?
* When did you feel something was wrong?
* Did you think you were having a stroke from your initial symptoms?
* When did you decide you needed to go to the hospital?
* Who was with you when you decided to come to the hospital?
* Do you have a history of stroke or any prior knowledge of stroke?
* What type of treatment did you expect to get at the hospital?
* Did you feel any urgency to get to the hospital quickly?
* Did you know of any treatments available for stroke patients?
During the interview, the patient was continually assessed for signs of fatigue. If a family member was present during the decision‐making process, the caregiver also was interviewed whenever possible to ensure the patient did not tire during the interview.
Grounded theory was used to discover the conceptual themes. Glaser and Strauss (1967) identified the following four stages in the constant comparative method:
1. Establish categories based on similarity of content in incidents and dissimilarity of content with other categories, with the aim of elucidating the theoretical properties of each category.
2. Compare each incident within each category with the dimensions of the category for integration into a unified whole that reflects the relationship of the dimensions or properties of the category.
3. Examine categories and their properties for underlying uniformities that may reduce the number of categories. Look for theoretical saturation of content.
4. Produce analytic memos to summarize the theoretical explanations.
Data were coded line by line as they were examined to identify and name concepts. The concepts were categorized, recategorized, and condensed into themes. This process continued until saturation was reached. After 10 patients or caregivers were interviewed, the themes that emerged were recurrent and consistent, with the previous themes identified as an indication of data saturation (Morse & Field, 1995).
Ten participants were interviewed; 5 were patients and 3 were caregivers; two interviews included both the patient and his or her caregiver. The average patient age was 65.6 years (SD = 13), with a range of 41‐82 years. Six patients were men and 4 were women. Three lived alone, but only 2 of these 3 participants were alone at the time of their stroke. Five patients lived 0‐5 miles from the hospital, 3 lived 5‐10 miles away, and 2 lived more than 20 miles away. The mean length of time from onset of symptoms to ED arrival was 5.6 hours, with a range of 45 minutes to 2 weeks. Five participants presented early (within the 3‐hour treatment window), with a mean time of 80 minutes. All of the patients who presented early for treatment were not alone at the time of their strokes. Five participants presented late (outside of the 3‐hour window), with a mean time of 81.6 hours. Of those who presented early, 1 was enrolled in a neuroprotective clinical trial, 1 received rt‐PA, and the others had preexisting conditions or treatments that precluded the administration of rt‐PA.
The average NIHSS score was 10.2 (SD = 5), with a range from 5 to 20. For the patients who presented early, the average NIHSS score was 9.4. The average NIHSS score for patients who presented late for treatment was 11.0. Eight patients had symptoms consistent with a right‐hemispheric stroke, and 2 had symptoms consistent with a left‐hemispheric stroke. Seven patients arrived at the ED by ambulance, and 3 were transported in private vehicles. Two of the 3 patients who arrived by private vehicle first had been evaluated in a physician's office and instructed to come to the ED. Seven patients had a history of stroke or transient ischemic attack, and 5 had a family history of stroke. Table 1 features sample demographics according to early (3 or fewer hours after symptom onset) or late (more than 3 hours) presentation.
The patients or caregivers all gave detailed accounts of their experiences and the factors influencing their decisions to seek medical care. The severity of the strokes varied among patients, as well as the situations surrounding each experience. Themes emerged that were common to both early and late presenters; distinctive themes also emerged for the individual groups.
In both groups of patients, two common themes emerged: the severity of symptoms and past experiences.
Severity of symptoms
Severity of symptoms was expressed in different ways depending on a patient's condition. The presentation was based on how patients perceived their symptoms. Two of the participants who were alone at the onset of symptoms experienced stroke symptoms that left them unable to phone for help.
Participants who perceived more severe symptoms had a greater sense of urgency to go to the hospital. Two patients had symptoms of a stroke that were minor and did not present until the symptoms worsened in severity. For example, 1 patient had a headache for 2 weeks before hospitalization, but did not present to the hospital until her symptoms progressed. Another patient was having periods of confusion, but she did not have a sense of urgency to get to the hospital until she collapsed on the floor.
Although past experiences varied, they appeared to influence the decision to seek care. One late presenter reported that her mother had a history of stroke and said, “I knew my mother had them, but I never knew what they did for her. Sometimes she never even went until later and found out she had one.”
Another patient's wife said she did not feel any sense of urgency to get to the hospital. She stated, “I've seen it happen before.”
Past experiences also were discussed, with some frustration in terms of treatment options. One patient said, “I was already told the medication I need does not mix with the one I am on, so why come?” A caregiver described a patient's first stroke, which occurred when she was taking heparin for left‐arm numbness; her symptoms worsened, and she “could not have the clot‐busting drug because she was on heparin.” The caregiver continued to describe his frustration with the current stroke, saying “And now they [the hospital] are telling me she cannot have it [rt‐PA] because the one blood level is elevated because of her warfarin.”
Five themes emerged from patients who presented late for treatment: lack of knowledge, fear of hospitals, denial, living alone/symptoms occurred at night, and preexisting health concerns.
Lack of knowledge
This theme encompassed both a general lack of knowledge concerning stroke signs and symptoms and the urgent nature of stroke and its treatments. Patients who presented late were not aware of what was happening, did not know their symptoms were those of a stroke, or did not realize that treatment was available if they presented to the ED within the 3‐hour window.
One patient had a headache for 2 weeks before her stroke and had an episode at work during which she fell on the floor, was unable to move her left side, and had slurred speech. The episode lasted for 2 hours before she “could function again.” She said, “I thought ladies could not have them.” Another patient also had initial symptoms that were not recognized as a stroke. Her husband reported “confusion” off and on for a day; even when her symptoms were more severe, he thought she was having a heart attack.
Another patient had serious symptoms for 2 weeks. He described a severe headache, left‐sided weakness and facial droop, and slurred speech. The patient's wife said she knew it was a stroke, but believed, “It will clear up, it will go away.” The patient had a history of stroke and transient ischemic attacks that in the past had resolved without any significant deficits. The patient's wife assumed this always would be the case.
This theme also included a lack of knowledge regarding available treatments. Patients were not aware of what would happen to them when they arrived at the hospital. One patient said, “I thought I was coming for help, but did not know what would happen.” Another patient said, “I thought I would get one little test and a pill when I got here.”
Fear of hospitals
Two of the patients who presented late for treatment also expressed fear of hospitals as a reason to not seek treatment. The patients wanted to stay at home as long as possible or preferred to be evaluated in the doctor's office. Two patients initially were evaluated in a physician's office and then were referred to the ED. Two patients who feared hospitals said:
“I would not have come to the hospital if not coerced by my sister.”
“I waited until Monday until a doctor was in because I was not going to the hospital.”
The patients who demonstrated denial believed if they ignored the symptoms, they would get better. This resulted in 2 patients waiting up to 2 weeks to seek treatment for their symptoms. This theme was illustrated by the following quote:
“If I ignored it, it would go away.”
Another patient reported, “I thought it would go away like it did before.” When asked how quickly his last symptoms subsided, he stated, “Within a few hours; I guess 2 weeks was pushing it.”
Living alone/Symptoms occurred at night
Two patients who lived alone experienced onset of symptoms when they were alone. Another patient was aware he was having a stroke and presented soon after he awakened, but he was not a candidate for treatment because he awoke with symptoms, which put his presentation outside of the 3‐hour treatment window.
Preexisting health concerns
Preexisting health concerns were unique in two cases in which minor stroke symptoms were mistaken for preexisting conditions. One patient had increased weakness in her legs that was attributed to her arthritis. Another patient attributed his symptoms of weakness in his left leg to his knee problems.
The 5 participants who presented early for treatment had a previous history of stroke or knowledge of stroke symptoms. This included knowledge of symptoms and an awareness of treatment options.
Previous history or knowledge of stroke
This theme was almost exclusive to the patients who presented early. These patients were aware of what was happening to them and knew to get help immediately. Not all of these patients had a history of stroke, but they knew about stroke or had a family member with a history of stroke. The participants described their knowledge level or awareness in various ways:
“My husband had a stroke 1 month ago, so I knew what was happening.”
“I read articles about strokes and knew what was happening.”
The patients who were more aware of what was happening were also more aware of available treatments. One patient was aware of rt‐PA and knew he could not receive it because he took warfarin. Other participants made the following comments:
“I was upset I did not receive the clot‐busting drug.”
“I wanted to get to the hospital as soon as possible so they could lower my blood pressure.”
Categories Based on the HBM
The themes that emerged from the caregivers and participants appear to support the HBM as the theoretical underpinning for understanding health‐seeking behaviors. For this study, these behaviors comprised the decision‐making process for seeking treatment for stroke. The themes that emerged from the data can be grouped into categories of perceived susceptibility, perceived seriousness, perceived benefit, and perceived barriers from the HBM. The categories are further divided into patients who presented early and late for treatment. The model illustrates the categories with the related themes (Fig 1).
The average length of time from onset of symptoms to ED arrival was 45.6 hours, and ranged from 45 minutes to 2 weeks. Five of the 10 patients presented within the 3‐hour treatment window. Themes identified by patients who presented late for treatment were lack of knowledge, fear of hospitals, denial, living alone/symptoms occurred at night, and preexisting health conditions. Those who presented within the 3‐hour time frame had knowledge of stroke from a previous history of stroke (self or family member) and/or were aware that treatments were available. Some patients were aware of “clot‐busting drugs,” and many patients wanted “help.” Both groups discussed severity of symptoms and past experiences as influencing their decision of when to seek treatment.
The participants who had previous knowledge of stroke symptoms responded more quickly than patients who did not have knowledge of the symptoms, with the exception of 2 patients who had such severe symptoms they were unable to call for help. These participants had a much higher sense of urgency to get to the hospital as opposed to participants who did not view themselves as susceptible to stroke. This theme was consistent with a study conducted by Kothari and colleagues (1997), in which researchers found that almost 40% of patients did not know the signs, symptoms, and risk factors of stroke. The lack of knowledge about stroke risk factors was a disconcerting finding in this study. Nine of 10 patients had at least one risk factor for stroke, but no patient believed he or she was at risk for stroke. Only 2 participants made a reference to their past medical problems. One patient expressed concern that she wanted her blood pressure lowered “because that is always what gets me into trouble.” Another patient said, “You never know when the big one is coming, so I get here right away all the time.”
Perceived severity or seriousness of symptoms also influenced the decision to seek timely stroke treatment. Williams and colleagues (1997) concluded that most stroke patients are unaware of the warning signs of stroke and present late because they misjudge the seriousness of their symptoms. This category was closely related to the theme lack of knowledge. Participants, who had a knowledge base regarding stroke and stroke treatments or a previous history perceived their condition as more serious and had a greater sense of urgency to seek treatment. The urgency for treatment also was influenced by the severity of symptoms. Several patients started with symptoms that did not greatly interfere with their activities of daily living. These symptoms were not recognized as those of a stroke or viewed as serious. They did not seek care until symptoms increased in severity and were perceived as more serious.
The lack of knowledge theme also was evident in the way patients accessed the medical system. Those who responded quickly or viewed their symptoms as serious called 911 or immediately went to the hospital. Patients who did not view their symptoms as serious came by private vehicle to the hospital, although some of their symptoms were much more serious than those of other patients, yet were not perceived as such. Denial also was critical to this theme; among the 3 participants who were identified as experiencing denial, all presented outside of the 3‐hour time window.
The theme preexisting health concerns was a major influence on concern and treatment‐seeking behavior. Stroke can present with a bewildering array of symptoms, and new symptoms of acute stroke mistakenly can be interpreted as related to preexisting health problems. Smith and colleagues (1998) found that patients with preexisting health concerns also had longer delay times.
One interesting finding was that there seemed to be no association between NIHSS score and presentation time. It would seem that the higher the NIHSS score (indicating a more severe stroke), the earlier the presentation. In this small sample, this was not the case. Several patients who had low NIHSS scores responded immediately, whereas other patients with more severe symptoms responded much later.
Another interesting finding was that all of the patients under the age of 60 years responded late. The sample size was not large enough to draw any conclusions; however, several of these patients or caregivers commented, “I am too young to have a stroke.”
The main conclusions of this small study are related to a lack of knowledge regarding stroke. According to the findings, a lack of knowledge persists regarding signs and symptoms of stroke, especially if initial symptoms are not severe. There also is a lack of knowledge regarding stroke risk factors and treatment for stroke. Both of these findings have profound implications for practice and should provide direction for future interventions regarding stroke education. It is crucial that educational programs include information on risk factors, stroke signs and symptoms, and treatment.
This study also found that living alone or experiencing symptoms at night dramatically altered treatment options and outcomes because all three patients for whom this was the case presented outside of the 3‐hour treatment window.
The major limitation of this study was the small sample size, which decreased generalizability. Although saturation was reached with recurrent themes within the interviews, diverse sampling was not possible (e.g., various ethnic groups, distance from the hospital).
Another limitation was that several of the participants interviewed were caregivers because some patients were incapable of completing an interview. Some factors such as fear and denial may not have been captured during caregiver interviews. Also, due to the study design, it could not be determined if others had influenced patient decisions to seek treatment.
It could not be determined from the sample if the patients who delayed treatment were unaware of their symptoms as a result of cognitive impairment. Anosognosia, or the loss of recognition or awareness of a hemiplegia, has been documented as a reason for which patients delay seeking treatment (Heilman, Barrett, & Adair, 1998). The influence of anosognosia on treatment delays could not be evaluated in this study due to the design. The presence or absence of anosognosia was not determined at the time of data collection.
Implications for Nursing Practice and Research
This study identified several implications for nursing practice. Nurses must take a leading role in educating the public regarding stroke signs and symptoms, risk factors, available treatments, and the concept that “time is brain.” An educated public is vital in improving stroke outcomes and the time frame in seeking medical care. In this small sample, 50% of the patients who presented early for treatment either had direct knowledge about stroke or a caregiver/spouse who had previous knowledge about stroke. Medical research promises many new future treatments, but all are based upon receiving treatment within a narrow treatment window. Hospitals may have the latest technology and treatments available, but these are of no avail if the public does not recognize the seriousness of stroke and the need for immediate medical attention. Further, educational efforts must be directed to the community as a whole rather than only those at high risk for stroke because the presence of other people influenced response time in this study.
It is estimated that 70% of strokes can be prevented. It is crucial that patients be aware of the risk factors for stroke and the importance of controlling risk factors, especially hypertension. It also is critical to stress that stroke can happen at any age. Everyone needs to be aware of stroke signs and symptoms, not just “older people.”
Although data saturation was reached with this small design, several additional areas for future research were identified but could not be evaluated due to the design of the study's small sample size. Response time in relation to NIHSS score is an important area for future research. There was not much difference between early and late presenters in this sample, but this could be due to the small sample size. Additional areas for future research are whether the severity of stroke or being alone affect response time and the concept of denial in seeking care.
In summary, stroke remains a leading cause of death in the United States. The current available treatment for stroke has a narrow time window. It is vital to identify factors that influence a patient's decision to seek medical treatment and to develop interventions to decrease treatment delays because “time is brain.”