Ischemic stroke is the most common form of stroke, accounting for 87% of all strokes, and each year, approximately 795,000 people experience a new or recurrent stroke. Of the stroke deaths that occur each year, 54% occur before the patient reaches the hospital, and on average, every 4 minutes, someone in the United States dies of a stroke. The most recent statistics available indicate that stroke accounts for 1 of every 18 deaths in the United States, positioning stroke as the fourth leading cause of death in the United States (Roger et al., 2011).
Recently, clinical trials have explored the efficacy of administering recombinant tissue plasminogen activator (rtPA) in an expanded treatment window of up to 4.5 hours. The European Cooperative Acute Stroke Study found that mortality did not differ significantly between participants enrolled in the 0–3-hour treatment window compared with participants enrolled in the 3–4.5-hour treatment window. These findings represented an important advance in the treatment of acute ischemic stroke (AIS) in providing evidence that rtPA can be given safely to carefully selected patients treated 3 to 4.5 hours after a stroke. Therefore, the time window for the treatment of AIS with rtPA has been expanded up to 4.5 hours, providing for increased treatment opportunities (del Zoppo, Saver, Jauch, & Adams, 2009).
Patient delay in seeking treatment for stroke symptoms is the major factor limiting the delivery of definitive treatment for this condition. Despite decades of research focused on decreasing delay times, the majority of patients still do not seek treatment in a timely manner. This pilot study sought to identify the relationship of somatic awareness and symptom attribution to treatment-seeking behaviors, specifically time to care. In addition, because the instruments used in this study were modified to add stroke-specific questions, this study served as the initial stage in the development of instruments measuring somatic awareness and symptom attribution in the stroke population. Using the Modified Somatic Perception Questionnaire (MSPQ) and Symptom Interpretation Questionnaire (SIQ) and employing a descriptive, correlational design, this study examined the relationship of factors to treatment-seeking behaviors that had not before been studied together in this population. Patient delay remains the major impediment to receiving timely treatment for stroke, resulting in morbidity and mortality that are substantially higher than is necessary, considering that effective treatment is available.
Two cognitive factors identified in the literature that were found to have an impact on care-seeking behaviors in other populations are somatic awareness and symptom attribution. The limited number of studies that have been conducted exploring the concept of somatic awareness and care-seeking behaviors involved patients with chronic illnesses such as chronic pain, sickle cell disease, and heart failure (Jurgens, 2006; Main, 1983; McCrae & Lumley, 1998). The only study identified in the literature that explored the concept of symptom attribution involved patients with stroke (Williams, Bruno, Rouch, & Marriott, 1997). There have been no studies conducted examining these variables together in patients with AIS. The purpose of this study was to describe the relationship of somatic awareness and symptom attribution to treatment-seeking behaviors in individuals experiencing an AIS. The central research question of this study asked whether an individual who is somatically aware and is able to correctly attribute his/her symptoms to stroke will seek care within 4.5 hours of symptom onset, thus decreasing the morbidity and mortality due to stroke.
For purposes of this study, somatic awareness is defined as the ability to identify bodily sensations, as a sensitivity to physical sensation, and as a bodily activity secondary to physiological change (Frasure-Smith, 1987; Jurgens, 2006; Kenyon, Ketterer, Gheorghiade, & Goldstein, 1991; Main, 1983). It can be an awareness or perception of physical symptoms or sensations (Pennebaker & Epstein, 1983) and may be one of the most prominent components of the appraisal of illness (Croyle, 1992). Pennebaker (1982) defined a physical symptom as a signal that guides behavior and is assumed to be an accurate representation of the underlying physiologic state. However, bodily sensations are often vague, ambiguous, and subject to a variety of interpretations, leading to incongruity between what is actually occurring and what one perceives is happening. This perception or awareness of a physiologic state is dependent on internal and external cues as well as the beliefs that are held regarding what the symptoms represent. People “sift” through sensations and selectively attend only to those that confirm their expectations about health and illness, often ignoring or discarding input that contradicts these beliefs. As a result, some symptoms are amplified and others are ignored (Pennebaker, 1982).
According to Bakal, Steiert, Coll, and Schaefer (2006), somatic awareness involves the direction of attention to bodily experience and associated feelings for the purpose of achieving health. In a more recent article focusing on the role of somatic awareness in the care of patients with body distress symptoms, somatic awareness was used to guide changes in disease/illness beliefs, body schema, coping styles, interpersonal dynamics, sleep, and medication dependence (Bakal, Coll, & Schaefer, 2008). Kirmayer and Robbins (1994) emphasize that the roots of attribution lie in bodily and social processes that are highly dependent on context and must be understood as part of the construction of a local world of meaning. Attributional styles can be viewed as one’s reflection of the underlying schemata that are used to interpret, label, and express new bodily and emotional sensations (Robbins & Kirmayer, 1991). Attributions concerning the cause of common somatic symptoms may influence later help seeking and illness presentation (Bower, West, Tylee, & Hann, 2000).
According to Ryan and Zerwic (2003), the interpretation of illness is a multidimensional process, involving assessment, information processing, and decision making. Symptoms are classified by the individual in terms of cultural perspectives and a hypothesis developed as to their cause. On the basis of individual attitudes and beliefs, a person makes decisions on what to do about his/her symptoms (Ryan & Zerwic, 2003). Senesky (1997) posits that patients’ responses to physical symptoms and to diagnosed illness are influenced substantially by their beliefs and attitudes. Attributions are among the cognitive constructs making up the individual’s own model of symptoms and illness (Senesky, 1997). Thus, the understanding of these individual differences in symptom identification is fundamental to developing tailored public information campaigns designed to shorten time to hospital presentation. Increasing accurate symptom recognition among high-risk individuals in the community may play a role in decreasing the time to hospital presentation after the onset of stroke symptoms (Williams, Rosamond, & Morris, 2000).
Review of the Literature
In a qualitative study attempting to understand treatment-seeking delay in women with acute myocardial infarction (AMI), Rosenfeld et al. (2005) identified six common themes: knowing and going, knowing and letting someone else take over, knowing and going on the patient’s own terms, knowing and waiting, managing an alternative hypothesis, and minimizing. A study involving patients with chest pain found that those patients who believed they were having a heart attack sought help more quickly than did those who did not (O’Carroll, Smith, Grubb, Fox, & Masterton, 2001). Moser, McKinley, Drakup, and Chung (2005) found that the factors that had a differential effect on delay between men and women who had experienced an AMI included age, history of AMI, type of AMI, concerns regarding not wanting to bother others, and prior knowledge of thrombolytics. In a population-based national cohort study (Howard et al., 2008), socioeconomic status (income or education) was associated with an increased likelihood of seeking care. Factors associated with a decreased likelihood of seeking care were individuals with previous stroke symptoms, without a history of heart disease, and without diabetes and past smokers (Howard et al., 2008).
Studies conducted in this country have involved the exploration of somatic awareness and care-seeking behaviors in patients with coronary artery disease (CAD), myocardial infarction, chronic pain, heart failure, and sickle cell disease (SCD; Jurgens, 2006; McCrae & Lumley, 1998; Warner, 1995). Findings in these studies were mixed. In a study involving patients with heart failure, an increase in the heart failure somatic awareness score was positively correlated with an increased delay in care seeking and was also a significant predictor of duration of acute symptoms (delay; Jurgens, 2006). Somatic awareness was positively correlated with pain episode duration in a study involving patients with SCD (McCrae & Lumley, 1998). Warner (1995) found that the somatic awareness scores differed significantly between women with CAD (lower scores) and women without CAD (higher scores).
Clearly, there have been mixed results in these studies. It is important to note that the majority of the studies reviewed involved patients experiencing symptoms due to chronic illnesses versus patients experiencing symptoms of an acute illness, such as a stroke. In addition, somatic awareness has not been studied in the stroke population.
Few studies have been conducted that have examined the role of symptom attribution in association with prehospital delay in patients with stroke. Williams et al. (2000) found that stroke patients presenting to the emergency department (ED) who had attributed their symptoms to stroke tended to respond to their symptoms sooner (1.4 vs. 2.5 hours) and arrive in the ED faster (2.6 vs. 4.8 hours) than do stroke patients who attributed their symptoms to other conditions. Hsai et al. (2010) identified that only 12% of patients who had experienced a stroke had actually called 911 first, 75% of patients experiencing a stroke reported calling a friend or relative, and 89% of these patients reported significant delay in seeking medical attention. The reason for delay in almost half of the patients was thinking the symptoms were not serious and/or that these would resolve on their own (Hsai et al., 2010). A prospective, multicenter study conducted in Korea (Kim et al., 2011) found that early arrival within 3 hours of symptom onset was significantly associated with (1) previous stroke, (2) high National Institutes of Health Stroke Scale (NIHSS) score, (3) atrial fibrillation, (4) use of an ambulance, and (5) knowledge about thrombolysis and awareness by the patient/bystander that the initial symptom was a stroke. In a study of patients with a diagnosis of ischemic stroke (Mandelzweig, Goldbourt, Boyko, & Tanne, 2006), perceiving control over symptoms (feeling able to control symptoms to some extent), attributing symptoms to problems other than stroke, hesitations about seeking help (verbalization of apprehension about seeking medical help), and an initial reaction of self-treatment or waiting to see what happens were associated with longer time intervals to both end points.
Design and Methodology
Recruitment took place in three acute care hospitals located in Philadelphia, Pennsylvania, and Binghamton, New York.
A clinical nurse specialist and nurse practitioner functioned as an honest broker within each hospital to assist in facilitating the identification and recruitment of participants for this study. Each honest broker screened potential participants according to the inclusion criteria, provided a patient letter explaining the study, and, if the patient indicated that he/she would like to participate, contacted the primary investigator, who then met with the potential participant. The primary investigator did not have access to the medical record.
Twenty participants diagnosed with AIS were recruited for this study during their inpatient stay within 7–10 days of admission.
Patients were evaluated for inclusion by the following criteria: (1) neurological status determined by a grade of ≤13 on the NIHSS at baseline, (2) diagnosis of ischemic stroke, (3) 18 years or older, (4) Mini-Mental State Examination score of >24, (5) may or may not have received tissue plasminogen activator (tPA), and (6) able to read, speak, and understand English.
A descriptive correlational design was used in this pilot study to explore the relationship of somatic awareness and symptom attribution with treatment-seeking behaviors in persons experiencing an AIS.
Time to care was operationalized as the number of hours/minutes from symptom onset to arrival at the ED. Symptom onset is defined as the time when signs or symptoms of the main complaint began (Evenson, Rosamond, Vallee, & Morris, 2001). Somatic awareness was measured using the MSPQ. The MSPQ is a valid and reliable 13-item questionnaire that uses a 4-point Likert scale to measure general somatic awareness (Table 1). Scores range from 0 to 39, with higher scores indicating higher levels of somatic or bodily awareness (Main, 1983). Because the MSPQ has not been previously used to study stroke patients, seven questions specific to stroke symptoms were added to the questionnaire. The seven questions asked whether the patient had experienced any of the following: sudden numbness/weakness of the face, arm, or leg, especially on one side of the body; sudden confusion; sudden trouble speaking or understanding; sudden trouble seeing in one or both eyes; sudden trouble walking; loss of balance or coordination; sudden dizziness; or sudden severe headache with no known cause. These questions were reviewed for content validity by three experts in the field and by three individuals who had previously suffered a stroke to determine whether the questions were readable and could be understood by the participants prior to inclusion in the questionnaire.
Symptom attribution was measured using the SIQ, a general measure of symptom attribution. It contains 13 common somatic symptoms, each with a set of three possible attributions or explanations: (1) emotional distress, (2) physical illness, and (3) normal environmental causes. Participants indicate on a 4-point Likert scale the extent to which they attribute each of the 13 symptoms to each potential cause (Table 1). Responses are summed to generate scores for each attribution, ranging from 13 to 52 (Robbins & Kirmayer, 1991). Because the SIQ is a general measure of symptom attribution, seven questions specific to stroke symptoms were added to the questionnaire. The seven questions asked whether the patient had experienced any of the following: sudden numbness/weakness of the face, arm, or leg, especially on one side of the body; sudden confusion; sudden trouble speaking or understanding; sudden trouble seeing in one or both eyes; sudden trouble walking; loss of balance or coordination; sudden dizziness; or sudden severe headache with no known cause. These questions were reviewed for content validity by three experts in the field and by three individuals who had previously suffered a stroke to determine whether the questions were readable and could be understood by the participants prior to inclusion in the questionnaire.
Correlation analysis was used to assess and demonstrate the relationship of somatic awareness and symptom attribution with treatment-seeking behaviors. Descriptive statistics, including mean, median, standard deviation, and graphs, were run for each continuous variable to understand the distribution of the variables. Frequency distributions were used to analyze the demographic data collected. The a priori significance level was set at .10 because this was a pilot study with a small sample size.
Table 2 shows that the final sample of 20 patients was composed of the same number of female and male patients and was predominantly white. Mean age was 62.84 years with a standard deviation of 13.17 years. Participants ranged in age from 41 to 92 years. Five participants had a personal medical history of diabetes mellitus (DM) and 10 participants had a personal medical history of hypertension. Seven participants indicated that they had a family history of stroke. Marital status was almost evenly split within the sample and most patients were educated at a high school level or greater. Three patients had no insurance, whereas the majority of patients had private insurance and/or Medicare.
Time to Care
Six patients in the sample did not remember their time of arrival to the ED; therefore, total time to care for these participants could not be calculated. The mean total time to care for the remaining participants was 851.43 minutes (or 14.2 hours) with a standard deviation of 1,400.29 minutes. The number of minutes that it took participants to arrive at the ED from onset of symptoms ranged from 30 to 4,560 minutes. Of the 21 participants, 12 recalled that they experienced symptom onset between 3:00 and 8:30 a.m., with the majority occurring between the hours of 3:00 and 7:30 a.m. The remaining participants, with the exception of four who could not remember, indicated that their onset of symptoms occurred during the late morning and evening hours.
The mean score for the MSPQ in the study sample was 14.00. Median was 12.00 with a range of 29 and a standard deviation of 8.176. Although there was a moderate correlation (r = .382) of somatic awareness to time to treatment, this was not significant (p = .178). Higher somatic awareness was not associated with the time interval between onset of symptoms and the presentation for treatment. Pearson’s product–moment correlation coefficient r for normal distribution was utilized to look at the correlation between the predictor and outcome variables (Table 3).
Significant correlation (r = .386, p = .093) was found between somatic awareness and education. In addition, t-test comparisons of group means to somatic awareness were completed, looking at the descriptive variables of gender and a personal medical history of hypertension. There were no significant findings to report with these comparisons. Because only five participants had a personal medical history of DM, this small number precluded additional t-test analysis of this variable to the outcome variables. With the risk of stroke tripled in patients with DM, it would be important to look at these comparisons in future studies involving somatic awareness and symptom attribution in ischemic stroke patients. Despite the lack of significance, in a comparison of group means between men and women, the mean somatic awareness score for the female participants was slightly higher than that of the male participants.
The mean SIQ score for the study sample was 106.85. Median was 110.00 with a range of 92 and a standard deviation of 24.618. Although there was a moderate correlation (r = .340) of symptom attribution to time to treatment, this was not significant (p = .234). Correct attribution of symptoms to a physical illness was not associated with the time interval between onset of symptoms and the presentation for treatment. Pearson’s product–moment correlation coefficient r for normal distribution was utilized to look at the correlation between the predictor and outcome variables (Table 3).
Significant correlation (r = .433, p = .057) was found between symptom attribution and education. In a comparison of group means between marital status and SIQ scores using analysis of variance/post hoc testing, there was no significance identified. In addition, t-test comparisons of group means to symptom attribution scores were completed, looking at the descriptive variables of gender and a personal medical history of hypertension. There were no significant findings to report with these comparisons. Because only five participants had a personal medical history of DM, this small number precluded additional t-test analysis of this variable to the outcome variables. With the risk of stroke tripled in patients with DM, it would be important to look at these comparisons in future studies involving somatic awareness and symptom attribution in ischemic stroke patients. Despite the lack of significance, in a comparison of group means between men and women, the mean symptom attribution score for the male participants was slightly higher than that of the female participants in the sample.
The SIQ contains 13 common somatic symptoms, each with a set of three possible attributions: (1) emotional distress, (2) physical illness, and (3) normal environmental causes. Participants indicate the extent to which they attribute each of the 13 symptoms to each potential cause, which constitutes the total SIQ score. To determine if there was an association between physical attribution scores and total time to care, the physical attribution was calculated separately for each participant. The question was whether attributing symptoms to a physical illness such as a stroke was associated with a shorter total time to care.
The mean physical attribution score for the study sample was 35.53. Median was 31.00 with a range of 43 and a standard deviation of 12.885. In a comparison of the mean physical attribution scores between those participants with insurance and those without insurance, significance was identified (p = .038). The mean physical illness attribution score for those participants with insurance was 37.14 compared with 24.00 for those participants with no insurance. In addition, the mean physical attribution scores in patients with a family history of stroke were significantly different from those in patients with no family history of stroke (p = .026). The mean physical attribution score for those participants with a family history of stroke was 40.25 compared with a mean score of 34.40 for those participants with no family history of stroke.
In this pilot study, the reliability analysis of the modified MSPQ and SIQ revealed Cronbach’s alpha of .73 and .94 respectively. Additional analysis of the scale was performed, examining the Cronbach’s alpha if one of the added items was deleted. The Cronbach’s alpha was between .69 and .74 for the MSPQ and .93 and .94 for the SIQ. These findings support that further use of these modified tools by researchers studying somatic awareness and symptom attribution can be undertaken with some confidence.
Somatic awareness and symptom attribution were not found to be significantly correlated to time to care. These results must be tempered with the knowledge that the data were derived from a very small sample and that some data, critical to the analysis, were missing due to a lack of recall. Further analysis did identify significance in comparisons of the group means, involving the variables of level of education, insurance status, and family history of stroke, to somatic awareness and symptom attribution. Additional t-test comparisons of group means to symptom attribution scores were completed, looking at the descriptive variables of gender and a personal medical history of hypertension. There were no significant findings to report with these comparisons. Because only five participants had a personal medical history of DM, this small number precluded additional t-test analysis of this variable to the outcome variables. With the risk of stroke tripled in patients with DM, it would be important to look at these comparisons in future studies involving somatic awareness and symptom attribution in ischemic stroke patient.
Individuals with an increased level of education, health insurance, and a family history of stroke were more somatically aware and/or able to correctly attribute their symptoms to a stroke than were those with lesser education, no health insurance, or family history of stroke. This pilot study served as the initial step in the development of instruments to measure somatic awareness and symptom attribution in patients who have experienced a stroke. The reliability findings in this pilot study of the modified instruments provide support for further development of their use in measuring somatic awareness and symptom attribution in patients who have experienced a stroke.
Future research, including qualitative research, should be conducted to study these variables. During the interview process in this study, patients very much wanted to talk about their “lived experience” of having a stroke. The findings in this pilot study give reason to believe that further research is necessary to take the next step to having a better understanding of the relationship of these factors to treatment-seeking behaviors in the AIS patient. Some of the modifications to the methodology that should be considered include (1) expansion of the timeframe for recruitment, (2) increasing the types of settings in which recruitment could take place, (3) increasing the sample size, (4) surrogate participation, and (5) requesting access to the medical record. The implementation of some or all of these changes as part of revised methodology would help to eliminate some, if not all, of the limitations seen with this pilot study.
In conclusion, this pilot study is an example of the type of study that has been recommended by researchers to examine cognitive and social factors that may contribute to delay in seeking care. Improving patient outcomes from a stroke relies on a patient’s ability to be aware of physiological changes or symptoms that herald the onset of a stroke, to correctly attribute the symptoms to stroke, and to act upon these changes by seeking early care and treatment. It will be essential to consider individual attitudes and beliefs when developing interventions, such as public information campaigns, as these components are critical to individual decision making regarding what to do about the symptoms that are being experienced. Although this study did not identify a correlation between somatic awareness or symptom attribution and time to care, there was a relationship identified between somatic awareness, symptom attribution, and physical attribution scores and some of the demographic variables, which merits further investigation.
Improving patient outcomes from a stroke relies on a patient’s ability to be aware of physiological changes or symptoms (somatic awareness) that herald the onset of a stroke, to correctly attribute the symptoms to stroke (symptom attribution), and to act upon these changes by seeking early care and treatment. To provide an accurate representation of these factors in this study population, it was necessary to modify the measures used in this study, the MSPQ and SIQ, with stroke-specific questions. The MSPQ, originally developed for use in patients with chronic pain, measures an individual’s ability to be aware of a physiological change (Main, 1983). The SIQ measures the three dimensions of symptom attribution (psychological, physical, normalizing) felt to be important to illness behavior (Robbins & Kirmayer, 1991). That is, when an individual experiences a symptom(s), is it attributed to a psychological reason, such as stress or a physical illness such as a stroke, or considered normal for that individual, such as in the case of an individual who has a history of migraine headaches. The dimension to which an individual attributes the symptom(s) is important in determining whether the individual will take action or not. The reliability of these modified instruments in this study provides support for further development of their use in measuring somatic awareness and symptom attribution in patients who have experienced a stroke.
Although this study did not identify a correlation between somatic awareness or symptom attribution and time to care, there was a relationship identified between somatic awareness, symptom attribution and physical attribution scores, and some of the demographic variables, which merits further investigation. In addition, despite a lack of significant findings identified between the primary study variables, we would be remiss not to strongly consider conducting another study involving these variables, using both a larger sample size and a revised methodology. It is incumbent upon nurse researchers to continue to pursue knowledge regarding factors that impact treatment-seeking behaviors as well as to develop evidence-based interventions.
With the advent of an increasingly aged population and the growing numbers of people diagnosed with risk factors such as DM, hypertension, and hyperlipidemia, it speaks to the importance of focusing nursing research in the area of stroke. This pilot study is an example of the type of study that has been recommended to examine cognitive and social factors that may contribute to delay in seeking care. Further research examining these factors is needed and is increasingly important as we are living longer and, often, with comorbidities, which place us at increased risk for having a stroke. The ultimate goal of current and future interventions is to be able to effectuate significant improvement in morbidity and mortality due to stroke. Nursing, through research, innovation, and advocacy, is well positioned to play a major role in identifying areas in which to utilize current interventions and in the development of new evidence-based interventions.
The authors thank Fran Cornelius, PhD, clinical associate professor at Drexel University, and Janice Hinkle, PhD, associate professor at Catholic University of America, for all of their support in the completion of this study.
Bakal D., Coll P., Schaefer J. (2008). Somatic awareness in the clinical care of patients with body distress symptoms. Biopsychosocial Medicine, 2 (6), 1443–1447.
Bakal D., Steiert M., Coll P., Schaefer J. (2006). An experiential mind–body approach to the management of medically unexplained symptoms. Medical Hypotheses, 67, 1443–1447.
Bower P., West R., Tylee A., Hann M. (2000). Symptom attribution and the recognition of psychiatric morbidity. Journal of Psychosomatic Research, 48, 157–160.
Croyle R. (1992). Appraisal of health threats: Cognition, motivation, and social comparison. Cognitive Therapy and Research, 16 (2), 165–182.
del Zoppo G. J., Saver J. L., Jauch E. C., Adams H. P. (2009). Expansion of the time window for treatment of acute ischemic stroke with intravenous tissue plasminogen activator: A science advisory from the American Heart Association/American Stroke Association. Stroke, 40, 2945–2948. doi:10.1161/STROKEAHA.109.192535
Evenson K., Rosamond W., Vallee J., Morris D. (2001). Concordance of stroke symptom onset time: The Second Delay in Accessing Stroke Healthcare (DASH II) study. Annals of Epidemiology, 11, 202–207.
Frasure-Smith N. (1987). Levels of somatic awareness in relationship to angiographic findings. Journal of Psychosomatic Research, 31, 545–554.
Howard V., Lackland D., Lichtman J., McClure L., Howard G., Wagner L., Gomez C. (2008). Care seeking after stroke symptoms. Annals of Neurology, 63, 466–472.
Hsai A., Castle A., Wing J., Edwards D., Brown N., Higgins T., Wallace J. (2010). Understanding reasons for delay in seeking acute stroke care in an underserved urban population. Stroke, 42, 1697–1701. doi:10.1161/STROKEAHA.110.604736
Jurgens C. (2006). Somatic awareness, uncertainty, and delay in care-seeking in acute heart failure. Research in Nursing & Health, 29, 74–86.
Kenyon L. W., Ketterer M. W., Gheorghiade M., Goldstein S. (1991). Psychological factors related to prehospital delay during acute myocardial infarction. Circulation, 84, 1969–1976.
Kim Y., Park S., Bae H., Cho A., Cho Y., Han M., Yoon B. (2011). Stroke awareness decreases prehospital delay after acute ischemic stroke in Korea. BMC Neurology, 11, 1–8. doi:10.1186/1471-2377-11-2
Kirmayer L. J., Robbins J. M. (1994). Symptom attribution in cultural perspective. Canadian Journal of Psychiatry, 39, 584–595.
Main C. J. (1983). The Modified Somatic Perception Questionnaire (MSPQ). Journal of Psychosomatic Research, 27 (6), 503–514.
Mandelzweig L., Goldbourt U., Boyko V., Tanne D. (2006). Perceptual, social, and behavioral factors associated with delay in seeking medical care in patients with symptoms of acute stroke. Stroke, 37, 1248–1253. doi:10.1161/01.STR.0000217200.61167.39
McCrae J. D., Lumley M. A. (1998). Health status in sickle cell disease: Examining the roles of pain coping strategies, somatic awareness, and negative affectivity. Journal of Behavioral Medicine, 21 (1), 35–55.
Moser D., McKinley S., Dracup K., Chung M. (2005). Gender differences in reasons patients delay in seeking treatment for acute myocardial infarction symptoms. Patient Education and Counseling, 56, 45–54.
O’Carroll R. E., Smith K. B., Grubb N. R., Fox K. A. A., Masterton G. (2001). Psychological factors associated with delay in attending hospital following a myocardial infarction. Journal of Psychosomatic Research, 51 (4), 611–614.
Pennebaker J. (1982). The psychology of physical symptoms. New York, NY: Springer-Verlag.
Pennebaker J., Epstein D. (1983). Implicit psychophysiology: Effects of common beliefs and idiosyncratic physiological responses on symptom reporting. Journal of Personality, 51 (3), 468–496.
Robbins J., Kirmayer L. (1991). Attributions of common somatic symptoms. Psychological Medicine, 21, 1029–1045.
Roger V., Go A., Lloyd-Jones D., Benjamin E., Berry J., Borden W., Bravata D. (2011). Heart disease and stroke statistics—2012 update: A report from the American Heart Association. Circulation, 123, 1–195. doi:10.1161/CIR, 0b013e31823ac046
Rosenfeld A., Lindauer A., Dorney B. (2005). Understanding treatment-seeking delay in women with acute myocardial infarction: Descriptions of decision-making patterns. American Journal of Critical Care, 14 (4), 285–293.
Ryan C., Zerwic J. (2003). Perceptions of symptoms of myocardial infarction related to health care seeking behaviors in the elderly. Journal of Cardiovascular Nursing, 18 (3), 184–196.
Senesky T. (1997). Causal attributions in physical illness. Journal of Psychosomatic Research, 43 (6), 565–573.
Warner C. (1995). Somatic awareness and coronary artery disease in women with chest pain. Heart & Lung, 24 (6), 436–443.
Williams L., Bruno A., Rouch D., Marriott D. (1997). Stroke patients’ knowledge of stroke: Influence on time to presentation. Stroke, 28, 912–915.
Williams J., Rosamond W., Morris D. (2000). Stroke symptom attribution and time to emergency department arrival: The delay in accessing stroke healthcare study. Academic Emergency Medicine, 7 (1), 93–96.