Maintaining Integrity: Women and Treatment Seeking for the Symptoms of Potential Cardiac Illness

Qualitative Health Research Volume 18 Number 11 November 2008 1461-1476 © 2008 Sage Publications 10.1177/1049732308325824 http://qhr.sagepub.com hosted at http://online.sagepub.com

Sheila A. Turris British Columbia Institute of Technology, Burnaby, British Columbia, Canada

Joy L. Johnson University of British Columbia, Vancouver, British Columbia, Canada Cardiac disease is the number one killer of Canadian women. Therapy is highly time-dependent, yet treatment-seeking delay persists. The purpose of this grounded theory study was to explore how women seeking treatment for the symptoms of potential cardiac illness interpreted their symptoms, made decisions about seeking treatment, and understood experiences of care in the emergency department. Data were drawn from 100 hours of naturalistic observation, indepth interviews with 16 women who visited one of two urban emergency departments between June 2005 and June 2006, and three interviews with nurses. The basic social psychological process of maintaining integrity was identified, encompassing three distinct phases: “resisting disruption,” “suspending agency,” and “integrating experiences and knowledge.” The findings suggest that in the face of a threat to life, women are first and foremost striving to maintain both the integrity of their sense of self and their ability to carry out role responsibilities. Keywords: emergency room; grounded theory; health care; health care, decision making; health care, users’ experiences; heart health; illness and disease, experiences; illness and disease, responses; women’s health

C

ardiovascular disease is the leading cause of death in Canada (Canadian Institute of Health Information, 2006) and the number one killer of Canadian women (Statistics Canada, 2004). Cardiovascular disease is pervasive in North America. Although the probability that an individual will die from cancer is 22%, the probability that someone will die from a cardiovascular cause is 47% (American Heart Association, 2004). The effectiveness of therapies for cardiac disease is often highly time-dependent. If women are admitted to an emergency department (ED) and treated within 4 hours after the onset of cardiac symptoms, there is a good chance that heart muscle that would otherwise be

Authors’ Note: This research was supported by scholarships and awards to S. Turris from the Canadian Institutes of Health Research (CIHR), the Michael Smith Foundation for Health Research, the Nexus Research Unit at the University of British Columbia, and the Xi Eta Chapter of Sigma Theta Tau International; and through a CIHR investigator award to J. Johnson. In addition, we acknowledge the helpful comments of Drs. Pamela Ratner and Bonita Long. Finally, we wish to thank the individuals who participated in this study.

destroyed can be salvaged. Current practice standards for the treatment of a heart attack support a goal of 30 minutes for presentation in the ED to the administration of fibrinolytic therapy (door-to-needle time), and 90 minutes for presentation at the hospital to the inflation of an angioplasty balloon (door-to-balloon time; Antman et al., 2004). Increasingly, attention is being directed to a mandate of treatment within 1 hour of symptom onset, given the number of deaths that occur in the first hour (McKinley et al., 2004). After the onset of symptoms, delay in presenting for treatment of the symptoms of potential cardiac illness is therefore of concern for health care professionals. There is considerable debate in the health care literature about gender and cardiac illness. Lockyer (2002) argued that cardiac care is gendered, explaining that we are still struggling with the image that heart disease is a problem of White, affluent men. For example, there is a persistent belief held by both genders that men are more prone to have heart attacks than are women (Hammond, Salamonson, Davidson, Everett, & Andrew, 2007; Lefler & Bondy, 2004). As well, both genders perceive that women handle stress 1461

1462 Qualitative Health Research

more effectively than do men, and so the former group is viewed as being protected from heart attacks (Richards, Reid, & Watt, 2002). There is also some evidence to suggest that women are more likely to delay than men; however, contradictory evidence continues to be reported. Although some researchers have found that being female is associated with greater delay in treatment seeking (Banks & Dracup, 2007; Bowker et al., 2000; Finnegan et al., 2000; Goldberg, Gurwitz, & Gore, 1999), others have found that gender contributes no unique variance to delay time (e.g., Banks & Dracup, 2006; Moser, McKinley, Dracup, & Chung, 2005) or that being male is associated with greater delay (White & Johnson, 2000). There has also been debate about the extent to which women and men share a common experience of the symptoms of cardiac disease. For example, the current position of the Canadian Heart and Stroke Foundation (2008) is that women and men might report the experience of symptoms differently, and that for women, pain might be of a vague nature. The American Heart Association (AHA) Web site reports that the most common symptom reported by men and women is chest pain or pressure, but that women might more commonly experience other symptoms such as shortness of breath (SOB). In fact, increasingly, there is an understanding that up to 30% of people who have acute coronary syndrome (ACS) will never experience pain as a symptom (Cornado, Pope, Griffith, Beshansky, & Selder, 2004). Well described in the health care literature are the differences in cardiac health treatment and outcomes, depending on whether you are a man or a woman. For example, according to the AHA Statistics Committee (2007), 38% of women will die within a year of their first acute myocardial infarction (AMI). Only 25% of men will die in the same period. Are these differences due to biology or gender? Rogers et al. (2006) studied 9,218 individuals (62.4% male, 37.6% female), followed for 100 days post-op (coronary artery bypass graft (CABG) surgery. Women were significantly more likely to die (16.1% versus 9.8%). Ninety-six percent of the difference in death rates could be explained by the more frequent rate of infections in women (pneumonia, urinary tract infections [UTI]). Even after controlling for age, length of stay, and comorbidities, the relationship between sex and infection persisted. Was the outcome of differential death rates because of biological gender (e.g., female urethras are shorter than males, and so the former group is more likely to

get a UTI), or is the difference because of gender in that women may be touched more often by staff, leading to higher rates of infection? Similarly, Blomkalns et al. (2005) published a review of existing gender disparities in the diagnosis and treatment of ACS (non-ST elevation myocardial infarction and unstable angina). They reviewed national data from 35,875 cases and found that women (who comprised 41% of the sample) were more likely to die in hospital (5.6% vs. 4.3%) and to reinfarct (4% vs. 3.5%). They also reported that women were less likely than men to receive an electrocardiogram (ECG) within 10 minutes of arrival in the ED and were less likely to receive heparin, angiotensen converting enzyme inhibitor (ACEI), and glycoprotein inhibitors. Reimer, Gitt, Borersma, and Simmons (2006) reported that women are less likely to be referred for stress testing, even after multivariate adjustment. Women also received antiplatelet and statin therapies less often than men and were twice as likely to suffer death or repeated AMI in the year following the first heart attack. In Canada, women admitted to the hospital with a new heart attack are significantly more likely to die in the hospital within 30 days than are men, even after age and comorbid conditions are taken into account. Thirty-day mortality rates, post heart attack, are reported to be 16% greater for women than for men (Canadian Institute of Health Information, 2006). Woodfield et al. (1997) reported that being female was an independent predictor of 30day mortality post heart attack. Similarly, Moser and Dracup (1993) reported that a first heart attack is more likely to be fatal in women than in men. Lansky et al. (2005) reported that women with a heart attack who undergo primary angioplasty have higher rates of mortality and morbidity than do men. Although confounding factors such as age and severity of disease undoubtedly play a role, we do not yet fully understand the reasons for these persistent differences. For example, the role of biology in outcomes is not yet clearly explicated. Women’s comparatively poorer outcomes post angiography might be partially attributable to the smaller relative diameter of female coronary vessels, which make angiography more difficult to perform. It is important to remember that conflicting evidence exists with regard to the association between female sex and cardiac disease. For example, Berger, Sanborn, Sherman, and Brown (2006) reported analyzed data from 4,284 consecutive patients, 31% of

Turris, Johnson / Maintaining Integrity 1463

whom were female. They reported no survival difference in-hospital and a trend toward improved survival for women, 3 years postpercutaneous coronary intervention. Yet, they pointed out that many other studies have found that even after controlling for age and comorbid conditions, a survival disadvantage exists for women. Similarly, Cox, Bata, Gregor, Johnstone, and Wolf (2006) reported data from all AMI admitted to hospital in Halifax (1984-2001) and found no association between biological sex and 28-day mortality. When considering the sometimes conflicting evidence regarding gender, cardiac illness, and treatment outcomes, recall that much of the research is retrospective. As well, studies often take place at teaching centers. Teaching centers tend to attract faculty and staff who are aware of current research—for example, about women’s rates of intervention and their outcomes when compared to men. And research has shown that expertise develops where many procedures are done. There is much debate in the health care literature about the influence of biological sex and gender on treatment-seeking behaviors. Treatment-seeking delay (TSD) is defined as “a prolonged interval of time from symptom onset to receipt of therapy” (Zerwic, 1999, p. 22). TSD is typically described in terms of patient delay, emergency medical services (EMS) delay, and hospital delay. Patient delay makes up the majority of the delay (Leitch, Birbara, Freedman, Wilcox, & Harris, 1989; Zerwic, 1999). To date, research has focused predominantly on describing the effects of individual factors such as the sociodemographic variables somewhat inconsistently associated with delay, such as sex, age, and past medical history (Ashton, 1999; Banks & Dracup, 2006; Bowker et al., 2000; Bunde & Martin, 2006; Dracup et al., 2003; Finnegan et al., 2000; Goldberg et al., 1999; Goldberg et al., 2002; Ho, Eisenberg, Litwin, Schaeffer, & Damon, 1989; Leitch et al., 1989; Lesneski & Morton, 2000; Moser et al., 2005; Pattenden, Watt, Lewin, & Stanford, 2002, 2003; White & Johnson, 2000; Wu, Zhang, Li, Hong, & Huang, 2004). Commonly, there is a hypothesis that women are more likely to delay than men. Despite more than 2 decades of research, contradictory evidence continues to be reported. Although some researchers have found that being female is associated with greater delay in treatment seeking (Bowker et al., 2000; Finnegan et al., 2000; Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto (GISSI), 1995; Goldberg et al., 1999; Lesneski & Morton, 2000;

Walsh, 1993), others have found that gender contributes no unique variance in delay time (Banks & Dracup, 2006; Moser et al., 2005; Zerwic, Ryan, DeVon, & Drell, 2003) or that being male is associated with greater delay (White & Johnson, 2000). Finnegan et al. (2000), in a gender analysis of TSD, reported that the women in their study delayed because (a) they attributed their symptoms to chronic conditions, (b) they had perceptions of low personal risk of AMI, and (c) they experienced higher levels of embarrassment related to false alarms. Of note, greater TSD among women runs counter to the overall tendency of women to access health care more frequently than men (Zerwic, 1999). Based on reports in the health care literature, gender can be said to be inconsistently related to TSD. When interpreting the results of research on gender and TSD, one must consider carefully the influence of population differences (by region), small sample sizes that provide insufficient power, and an absence of statistical control for confounding factors such as age and comorbidity. Recall that women are, on average, a decade older than men when they begin to experience symptoms suggestive of cardiac illness (Zerwic, 1999). And those who do not seek treatment at all are excluded from research about delay. It is unknown whether these numbers differ by gender. In the context of emergency services, little has been written in relation to gender and health care. However, O’Brian and Fothergill-Bourbonnais (2004) researched the experiences of trauma patients in the ED. They reported that the women in their study derived a sense of comfort from the presence of family and friends. In contrast, the men were less concerned about social support and were worried that their family and friends might “get in the way” of health care professionals if an emergent situation arose. Similarly, Redley, Beanland, and Botti (2003) reported that female gender, for both patients and family members, is associated with reports of greater anxiety during a visit to an ED (depending on age). And O’Donnell, Condell, Begley, and Fitzgerald (2005) reported that women experienced greater delays from “triage to medical assessment” than did men (30 minutes compared to 20 minutes), and greater delays in average door-to-needle time (70 minutes versus 52 minutes). Although this is a single study that took place in six major teaching hospitals in Ireland, the findings highlight the possibility that gender might influence outcomes.

1464 Qualitative Health Research

The process by which individuals make decisions about treatment seeking is complex and inevitably influenced by social factors. To gain a sense of this complexity we conducted a grounded theory (GT) study (Glaser, 1992) focused on understanding women’s experiences of, and decisions about, the symptoms of cardiac illness.

Method Grounded theory has its roots in sociology, pragmatism, and symbolic interactionism (SI). Accordingly, there is a focus on understanding the socially derived meanings that people attribute to events. Features include a focus on the development of theory that explains the processes that underlie social experience and social relations. Grounded theory has a holistic focus. Glaser (1998) suggested that researchers consider entire incidents rather than fragments of data. In the complicated world of health care, a sense of “what is going on here?”—with a focus on overall meaning, rather than discrete events—is invaluable for creating theory. The focus of GT is on understanding the basic social and/or psychological process that is taking place. We were interested in examining women’s experiences during an episode of symptoms suggestive of cardiac illness and explaining the process by which those women made sense of their experiences prior to, during, and after their visits to an ED. Women were chosen as participants for the study for three reasons. First, cardiac disease is the number one cause of mortality and morbidity for Canadian women (Statistics Canada, 2004). Second, as discussed in the literature review, there is evidence that women delay seeking treatment for symptoms of a cardiac event longer than their male counterparts (Banks & Dracup, 2007; Bowker et al., 2000; Finnegan et al., 2000; Goldberg et al., 1999), yet the reasons for this remain largely unexplored. Third, women are historically less likely to be participants in research studies designed to investigate cardiac disease (Bierman, 2007; White & Lockyer, 2001). The primary research question for this study was: In the context of the symptoms of a potential cardiac event, what is the process by which women understand and make decisions about those symptoms? Prior to undertaking this study, ethics approval was obtained from the university ethics board and the ethics boards of the two hospitals where data collection

took place. Data was collected from two EDs in a Canadian urban center, both with more than 50,000 ED visits each year. Data collection and analysis took place over one year, commencing in July of 2005. To increase the rigor of the study, more than one form of data was sought, including both naturalistic observation and open-ended interviews. Naturalistic observation took place at each of the sites on variable days of the week and on different shifts. Health care staff identified potential participants for the study, based on the chief concern as stated at the triage desk. If a woman met the inclusion criteria for the study (described below), she was then approached by the primary researcher, provided with a brief verbal introduction to the study, and then given a two-page letter that included contact information for the researchers. Initial questions about the study were answered at this point. If a woman expressed interest in participating in the study, permission was asked to telephone the individual, following discharge, to set up an interview at a time convenient for both the participant and the researcher. The advantage of this approach was that all of the participants had information about the study and time (e.g., more than 24 hours after hospital discharge) to consider whether or not they wished to participate in the study. We obtained the accounts of 10 participants as a starting point for this project and completed 80 hours of naturalistic observation. Theoretical sampling was employed to collect further accounts. For example, the initial data analysis highlighted the possible influence of age on decisions about treatment seeking, so we sought participants who were relatively younger to further examine the influence of age on decision making. Similarly, when the data analysis revealed themes about the form and function of the ED, we interviewed nurses regarding their perspectives about emergency services and the care they provided.

Sample and Recruitment To be eligible to participate in this study, women had to be visiting either of the two EDs for treatment of symptoms of a potential cardiac event, hemodynamically stable, 18 years of age or older, able to speak English, and competent to provide informed consent. Sampling of both participants and observations continued until the categories were dense and well developed and support for the relationships between the categories and the contributions of those categories to the basic social psychological process were clear. The

Turris, Johnson / Maintaining Integrity 1465

interviews took place between 2 and 6 weeks following discharge from the hospital. Upon the conclusion of the project, a total of 17 interviews had been completed with 16 women, an additional 3 interviews were completed with nurses, and field notes documented 100 hours of naturalistic observation.

Coding The heart of GT is a coding process that consists of open, selective, and theoretical coding. Each level of coding moves the data analysis toward increasing levels of abstraction. During open coding, the first step in the theoretical analysis, incidents were compared to other incidents, and similarities and differences were sought, compared with other concepts, and subsequently grouped to form categories. This process is termed “constant comparative analysis.” Open coding, beginning in the margins of the transcripts and field notes, involved making notes about the incidents and comparing those incidents to other incidents. Each incident was then coded into as many in vivo codes1 as possible. Initially, labels for the categories were drawn from the interviews directly. As the analysis proceeded, the labels became increasingly explanatory. Gradually, we grouped the information according to various relationships and interactions; important concepts began to become apparent in successive accounts. These concepts informed the development of the categories, which explained what was going on in the women’s accounts and our observations. Categories possessed two essential features: they were both analytic and sensitizing. The former refers to the quality of being sufficiently abstract and the latter refers to the ability of a code to generate a meaningful picture (Glaser & Strauss, 1999). Regular meetings of the research team assisted us in the process of data analysis; the discussions guided us to ask questions and to explore the relationships within and between the accounts of the participants, as the categories developed. Initially, our categories were general and broad; they became more specific and better developed as new data were constantly compared with that obtained in earlier interviews. Eventually, these categories were analyzed to examine the existing connections and, as some of the categories were collapsed and subsumed within others, we sought evidence of a core category. According to Glaser (1978), a GT study always has a core category that explains what is

occurring in the data. That core category might or might not be a basic social psychological process (BSPP; a process with phases that unfold over time). We ended open coding when a category was developed such that no new related concepts emerged. After open coding, we began selective coding. Selective coding was delimited to the data that were related to the core category that emerged (Glaser, 1992). The goal was to develop a core category that accounted for “most of the variation seen in a pattern” (p. 75). The core category developed as we coded the data at increasing levels of abstraction. The focus was on specifying a category in terms of the conditions that gave rise to it and on systematically integrating the relationships between the categories (Polit & Hungler, 1999). Selective coding guided the next phases of data collection and analysis. Finally, greater integration of the findings occurred. This took place through the process of theoretical coding—the process of returning to the data (or to the field) and using additional sampling to close the gaps in the data (Charmaz & Mitchell, 2000). As such, we conducted additional interviews and naturalistic observations. For example, the data analysis revealed what we termed the “discourse of deservedness”—that is, the notion that some individuals seeking care might be thought of as more deserving of the limited available resources than others, based on unwritten and unspoken criteria. To further explore this idea, we obtained the accounts of three nurses with expertise in emergency care.

Findings The ages of the participants ranged from 29 to 89 years. Slightly less than half were employed at the time of the interview. Roughly one quarter of the participants were married and half of the women had children. The majority of women experienced the onset of symptoms while at home. Seven of the participants reported that family members were present in the home when the symptoms began. In relation to the amount of time that elapsed between onset of the symptoms and treatment seeking, delay ranged from approximately 30 minutes to more than 48 hours. Twenty percent of the participants were transported to the ED by ambulance; 62% percent drove in a private vehicle. Roughly one third of the participants received a cardiac diagnosis prior to discharge.

1466 Qualitative Health Research

Although the focus of this study was women’s experiences of ED care, the women did not begin their stories with those accounts. Instead, without exception, the women began by talking about where they were when they first noticed their symptoms and how they experienced their symptoms. As their stories unfolded, the multitude of factors influencing the decisions they made in relation to seeking treatment became apparent. The experience of symptoms of potential cardiac illness differed for the women. For some the symptoms developed slowly, over a period of days, and for others the symptoms quickly escalated in severity. All of the women eventually recognized that they needed help and ended up visiting an ED. Common to all of the participants was an initial period of confusion about their symptoms, followed by a reasoned consideration of possible courses of action in relation to treatment seeking. The basic social psychological process of maintaining integrity explained the women’s actions and inactions during an episode of cardiac symptoms. Maintaining integrity refers to the process the women used to keep intact their sense of self and to keep going in the context of their daily lives, despite the threat represented by the symptoms of heart disease. Integrity had several dimensions including the personal, the social, and the physical. Personal integrity encompassed the women’s sense of themselves as capable, responsible, and worthy individuals. Social integrity was closely intertwined and captured the importance women attached to role responsibilities such as those related to being a “good” wife, mother, daughter, or employee. In contrast, physical integrity encompassed biological well-being and accordingly addressed the women’s images of themselves as healthy. The data from this study suggest that the participants moved through three phases, including “resisting disruption” (pre-hospital), “suspending agency” (in-hospital), and “integrating new knowledge and experience” (after discharge).

Resisting Disruption Resisting disruption described the women’s efforts to avoid any alterations to their daily routines (or those of their family members), as well as their efforts to preserve their physical well-being. In this phase, the women strove to maintain integrity by carrying on with their daily activities, in spite of the development of symptoms such as chest pain or shortness of breath. Resisting disruption had two stages, including “analysis of symptoms” and “managing the situation.”

Analysis of symptoms. Resisting disruption began with an analysis of the symptoms. The accounts of most of the participants began with a description of the development of awareness. Awareness involved taking notice that something out of the ordinary was occurring in their bodies. Sometimes this awareness took the form of recognizing a particular sensation, such as pain or dizziness. At other times, awareness that something was wrong arose from the inability to carry out normal tasks and routines such as showering or preparing a meal. This awareness resulted in ongoing self-monitoring, to track the duration and quality of these sensations. Whereas one might expect momentary twinges of pain or temporary bouts of weakness, when a symptom persisted for longer than a few minutes, their awareness and continued monitoring led to an acknowledgement that there was a threat to physical integrity at some level. Problem identification followed awareness and self-monitoring, and occurred when the physical sensation or inability to perform a task was judged to be more than a momentary aberration. At this time, the women began to interrogate their symptoms. Perception (awareness and problem identification) moved forward to comprehension (understanding). Assessing their general and personal risk for heart disease and looking for explanations occurred concurrently. The participants considered possible explanations for their symptoms, while simultaneously carrying out an assessment of their personal risk. Considerations about factors such as age and gender were often prominent, as were risk factors such as smoking, past medical history of cardiac disease, family history of cardiac disease, and stress level. Looking for explanations, another aspect of symptom analysis involved the consideration of a limited number of possible (usually three or fewer) explanations for the symptoms. Some of the women were concerned about the possibility of a cardiac cause; however, far more commonly, the women hypothesized that their symptoms were caused by dietary misadventures whether in the form of food poisoning or overindulgence. This was without question the most common explanation, even when the same food was consumed by another person without ill effect. In describing her search for explanations one woman commented, So, this will be explained by, I don’t know, too much coffee, too many cigarettes, or appalling lifestyle or . . .

Turris, Johnson / Maintaining Integrity 1467

I mean this will be it, you know, nothing serious is going to happen.

Stress was a common reason identified by the women as the root cause of their symptom(s). Stress was normalized and pervasive, serving as a ready explanation for the women’s experiences of symptoms. Labeling symptoms as “stress” gave women an explanation for their symptoms—an explanation that did not involve cardiac disease. Situational awareness, a final aspect of symptom analysis, involved projecting into the future and the predicting of outcomes. During symptom analysis, the participants defined the degree to which their symptom(s) were a threat to their personal, social, and physical integrity. If a woman chose (or avoided) a particular course of action, what were the possible outcomes? These outcomes were explicitly or implicitly ranked in relation to their desirability. As one woman explained, I really didn’t think it was anything . . . I was hoping it wasn’t. It was just that I thought I had to check it out. I just thought . . . all I could see was that if I went to work and something happened because there really was something and I was transferring somebody . . . and something happened and it would be my fault and I would be remiss and . . . I didn’t want to deal with that.

Managing the situation. Managing the situation was the second stage of resisting disruption. The women managed their situations in ways that allowed them to continue to address their role responsibilities and thus preserve their images of themselves as healthy, strong, and capable. Specifically, this stage included making decisions about how to respond to the symptoms as well as how to manage the role(s) that others might play, depending on the course of action chosen by a woman. Taking action, one aspect of managing the situation, involved a myriad of decisions and the subsequent initiation of a wide variety of strategies to manage the symptoms. Consulting others, including family members, coworkers, family physicians, friends, or the ambulance service was a common strategy used to gain an additional opinion about the possible causes of the symptoms, but more importantly to obtain advice about the best course(s) of action to take. There was often a time gap between the symptom analysis and the action. Several participants expressed a reluctance “to worry” their family or friends.

Self-treatment was also a popular course of action. Self-treatment was based on their symptom analysis and often consisted of trying medications, such as nitroglycerine or over-the-counter remedies such as antacids. An attempt to self-treat often delayed a visit to the hospital because the women waited to see whether their home interventions were effective. As well, the women often chose to watch and wait, monitoring their situations. This choice was motivated by a hope that their symptoms would be resolved without intervention. After the symptoms began, they considered two possible outcomes. First, there might be nothing wrong and they would be embarrassed about seeking treatment for a false alarm, not to mention the disruption of their lives that would be attached to a visit to an ED. Or there might be something wrong (something serious), in which case they were potentially very ill. When forced to choose between embarrassment and illness, watching and waiting was seen as a strategy to avoid either alternative. According to the accounts of the participants, avoiding embarrassment seemed to be more important than avoiding a serious outcome. Addressing the needs of others frequently preceded a visit to the ED and involved considering the needs of others, whether related to meal preparation, transportation to school, or work obligations. The women in this study did not want their loved ones to face long waiting periods in the hospital. Accordingly, the women often traveled to the hospital alone, having sent their partners to work for the day. As well, addressing the needs of others often translated into a time delay in getting to the ED. For example, one participant explained that although she had pain all night, she wanted to wait until the morning to seek medical attention because her partner worked 14-hour days and he needed his rest. He has to be at work at 5:30 so I think he dropped me off [at the ED] and I probably said, “It’s fine and I’ll let you know. I can phone and get you at [work], don’t worry about it.” Because it’s such a process. Like, he just sits around and waits anyway and they let him in after the fact, which could be an hour later or something. So no, he went to work and said phone him if . . . I just went in on my own.

Re-evaluating was also an aspect of managing the situation. New developments such as additional symptoms, the passage of a certain period of time (i.e., symptoms throughout the night and into the morning), or the failure of self-treatment caused the

1468 Qualitative Health Research

women to “take stock” and re-evaluate what was occurring. Re-evaluation always resulted in a reassessment of how significant the symptoms might be, causing the women to consider relatively more serious explanations for their symptoms. I don’t know whether that was it or what caused the ingestion. It just stayed with me . . . it was like a band around my middle, all into my back. My back hurt; it was almost as if I was having gallbladder attack because I remember how severe that was but I don’t have a gallbladder. It was a band around my stomach and it just stayed. It was sort of from the mid-chest down, not low and I thought, “Okay.” And I’d spent the whole night . . . I had taken Rolaids, I had taken Tums, then I had tried the Pepsi and that was it.

Before their admission to the hospital, the women viewed their symptoms as an unwelcome distraction to the business of everyday life. For them, maintaining personal, social, and physical integrity concerned addressing the expectations of others and preserving their sense of themselves as capable, strong, and healthy. For the participants, the ability to carry on with their daily routines in a private sphere was tightly linked to their sense of personal identity. The women generally defined themselves in relation to their social connections. That is, they defined themselves according to their responsibilities to the significant others in their lives. The participants shared particular values and beliefs. For example, putting the needs of others before oneself was a value commonly espoused by participants. This typically translated into a delay in seeking treatment. Accordingly, not awakening partners to express concern or to solicit a ride to the hospital was usually viewed as a more appropriate course of action than seeking medical attention promptly. Similarly, getting children off to school was ranked, by participants who had children, as more important than threats to personal health and the determination of the cause of symptoms. Thus, ideas about goodness and social/familial roles ultimately shaped and constrained the women’s actions so that the most attractive course of action often involved adopting an approach of watching and waiting, hoping that the symptoms would resolve spontaneously. In this way, many of the women in this study sought to maintain personal and social integrity by avoiding the consequences they perceived would be associated with failing to carry on with their daily lives. In other words, the important aspects of daily life commonly

crowded out the urgent requirement for action mandated by the experience of the symptoms of potential cardiac illness.

Suspending Agency Suspending agency, the second phase of maintaining integrity, occurred in the ED, a public sphere. Having tried to avoid disruptions, the women were forced by the nature of their evolving or persistent symptoms to reassess the threat represented. To this end, the women sought expert advice. During this phase, the women temporarily stopped trying to address role responsibilities related to employment or their roles as wives and mothers. They stopped trying to manage their symptoms at home, and they stopped trying to manage the actions of their significant others. The participants stopped active attempts to minimize disruptions to their daily lives. Instead, they entered into a relationship with their health care providers. Three stages were associated with suspending agency. The staff took control and the women experienced becoming dependent; the staff engaged in “constructing a cardiac case” and the women experienced “being repackaged;” and the staff made judgments about the patient’s deservedness and the women experienced being judged. The suspension of agency served several purposes for the women. Specifically, suspending agency was one way to ensure continued physical integrity. As well, temporarily setting aside the responsibilities and commitments of daily life was viewed as a trade-off— a loss of control in return for the resolution of the symptoms, or at least an explanation for the cause of their symptoms. Suspending agency was also a way for the women to manage their fear and anxiety. Having reassessed the threat level and determined that something serious might be causing their symptoms, the women reported emotional distress. Suspending agency and handing over control to the health care team for the most part reassured the women. Finally, suspending agency was an active strategy used by the women to support health care professionals’ ability to intervene. Therefore, docility, compliance, and cooperation were important features of this phase. Although suspending agency was a voluntary action, unique structural conditions within the ED supported, and indeed mandated, the surrender of control. The interplay between agency and structure was complex and the influence of structural conditions was multifaceted. Consider that before the ED visit, the experience of their symptoms was woman-centered

Turris, Johnson / Maintaining Integrity 1469

(e.g., the problem was defined by the woman, located in the woman’s experience of the symptoms, and she was the problem solver). Once in the ED, however, a pronounced shift occurred. The problem became defined by the health care professionals, the problem to be addressed was relocated in the individual case, and the problem solver(s) became the members of the health care team. The women entered the health care system as women, but moved through it as patients. During this phase, only infrequently did the women advocate for themselves, questioning treatment decisions and requesting information. Becoming dependent. The nursing and medical staff acted to take control from the moment the women presented to the ED. Oriented toward the treatment of acute illness, the staff used expert knowledge and experience to project the worst possible situation that might arise given a particular set of symptoms, and then took pre-emptive action to address that outcome. Working on the assumption that the worst-case scenario would occur, and defining all situations as potential emergencies, the health care staff adopted particular practices to gain rapid access to information about a woman’s health history and physical condition. Urgency and a bias toward the treatment of highacuity illnesses were embedded within the organization of space in the ED (in terms of layout, physical area, and patient flow), as well as the attitudes and practices of the health care staff. Thus, the efficient movement of patients within the department, as well as the expeditious management of emergency situations, were supported. Through their actions, the health care staff created the conditions necessary for dependence to develop and thrive. Women in this study did not have access to information and resources about cardiac illness during their visits to the ED, nor did they have access to a consistent source of information about their condition given shifting staff assignments. This created a knowledge vacuum for the women that had the indirect effect of reinforcing professional power, of causing persistent fear and anxiety for the participants, and of creating dependence on the health care professionals. Dependence served the interests of the health care professionals, ensuring that important health information was accessed expeditiously. Becoming dependent necessitated both disclosure and socialization to the ED culture. With the act of disclosing their symptoms, the participants surrendered the responsibility of solving the

problem. With the women in a dependent role, the ED staff could more easily take action and respond to the women’s symptoms in a predetermined way using predetermined strategies. The creation of dependence occurred as the women became socialized to the ED culture. Specifically, such socialization involved three things: drawing on past experience, learning about the ED environment, and cultivating qualities associated with being a “good” patient. Being repackaged. The experience of cardiac symptoms is personal and frightening. For health care professionals, the women’s experiences were something to be analyzed, measured, and categorized so that an accurate diagnosis might be reached. The women learned that their symptoms were being constructed and understood not in a personal context (e.g., a particular individual’s social, emotional, and physical world), but in the context of the accumulated biomedical research and knowledge about illness in general, and cardiac illness in particular. Interpreting symptoms in light of this knowledge required standardized approaches to care and particular behaviors on the part of the women. For example, to collect some of the data required for building a cardiac case, the woman had to be on a stretcher, lying down, undressed, and not moving. She had to be willing to undergo various, sometimes painful, procedures—often without explanation or discussion. And, if discussion did take place, it was often not reassuring. She also had to be willing to let people “practice” on her body, whether this was in the form of medical rounds or an IV start by a new paramedic. And she had to cooperate with being treated as a model or an object, rather than as a person. Nudity, exposure, pain, and lack of privacy were all features of the ED experience. The paramedic is poking me with a sharp instrument while he’s asking me all these questions so that was a little distracting but his questions . . . if I had to think of one word, looking back, I found him annoying . . . you know, he was just doing his job. But he could have either waited ‘til the guy stopped poking at me; it would have been more, I don’t know, tactful or something if he’d have just backed off for a few minutes until they finished what they were doing. So that was a little hard to answer his questions while that was going on.

To reach a diagnosis about the cause of the woman’s symptoms, the health care staff generated several hypotheses and then considered each possibility in turn.

1470 Qualitative Health Research

Constructing a cardiac case captured the approach health care professionals took in the assessment, care, and treatment of women with symptoms indicative of cardiac illness. In this context, repackaging signified paying attention to certain types of information. This information had to be obtained from each of the participants and served to assist the staff in standardizing the information they exchanged. For the women this translated into the experience of “surveillance.” Surveillance, in the form of cardiac monitoring, direct observation, and diagnostic tests, was the main method used to collect the information used to create a cardiac case. Alternatively, if the data could not be reassembled to fit with hallmarks of cardiac illness, they were used to dismiss the symptoms. Standardization and the need for surveillance data, to support the creation of a cardiac case, created several challenges for the women, particularly in relation to their experiences of ED care, and ultimately, the outcomes of that care. Breaches of confidentiality, objectification, and mistaken treatments were three obvious consequences. The cardiac pictures or cases generated by the staff served the interests of the women and their health care professionals because the cases could be used as a form of shorthand communication. Clinical pictures also were used by the ED nurses in gatekeeping and advocacy functions, and ultimately, to make management and resource allocation decisions. For example, if there was a single stretcher available and several patients in need of placement, two triage nurses would discuss each patient, pointing out atypical diagnostic findings and deciding which of the patients was most in need. ED staff members were acknowledged by the women in this study as having expertise and “knowing what to do.” For some of the participants, the “experts” were seen as all-knowing and powerful. One patient characterized her ED experience in this way: “It went really well. They just took over!” A belief in the professional’s expertise was so strong that one participant expressed dismay when she was consulted by a cardiologist. When he said, “Do you have any more questions (you think) I should ask you?” that was the most stupid thing! My specialty is not cardiology, but you know, that was the most stupid question anybody’s ever asked me . . . but he might have been nervous. I don’t think so. No . . . you’ve asked me everything that. . . . You’re the one! You’re asking the questions. What haven’t I told you?

The benefits of the application of expert knowledge were clear; however, there were unintended negative effects. For example, although classical cases (drawn from expert knowledge) functioned to alert the staff about when to take swift action, if a particular woman’s clinical presentation was atypical and did not fit the classical picture, cardiac causes might not be considered. For example, a 26-year-old woman who presented to the ED with chest pain and ECG changes indicative of cardiac ischemia described her experience of having her complaints dismissed by a nurse. “The nurses . . . the triage nurse, from my history, took it seriously. The other nurse, she came around and said, “You’re too young to have chest pains.” Being judged. The view that an individual was responsible for her health status was commonly held by both the staff and the participants themselves. “Healthism” was obvious in the ways that the women spoke about their risk factors, including obesity, smoking, diet, family history, compliance with physician advice, prescription regimens, and lack of exercise. For the staff, these factors were front-and-center when caring for their patients. On one occasion, for example, one of the staff nurses talked about her frustration that people wanted to smoke when she felt that they should be lying in their beds, recovering. She speculated that people can’t be really ill if they are well enough to go outside for a cigarette. Her expression and tone was one of distress and disgust. She acknowledged that of course smokers would want a cigarette, but questioned that smoking was an appropriate behavior when one was ill. The women were aware of the judgments by staff, and indeed they judged themselves in relation to their risk behaviors, as illustrated by one woman who, when speaking about her health history to a nurse, raised her hands in the air and stated emphatically, “OK, OK, I haven’t been taking my pills. I forget sometimes. Don’t judge me!” or, when another participant confided, “I don’t walk. I’m a bad girl!” One woman, who had in the past been employed as a nurse in an ED, was surprised not to be chastised, certain that she would be judged to have sought care inappropriately. He didn’t reprove me or anything like that and he felt sure that it was this radial nerve palsy and was good enough to do the blood work and listen to me. You know, it didn’t show up before without testing . . . without looking at the enzymes or whatever.

Turris, Johnson / Maintaining Integrity 1471

Integrating New Knowledge and Experiences The third phase of maintaining integrity was integrating new knowledge and experiences. This phase was focused on making sense of recent events. Once again in their home environments, the women reflected on their experiences of symptoms and of ED care, in the context of their diagnosis. At this point the women were striving to integrate new information while maintaining personal and social integrity, finding ways to incorporate new information while maintaining the status quo. In the third phase, it became apparent that experiences of care were not simply about understanding the character of the women’s interactions with the health care system via their visit to the ED, but also about the ways in which they constructed the meaning of those experiences. The construction of meaning continued after discharge as the participants made sense of their recent illness experiences. For the women who received a cardiac diagnosis, making sense of the problem was about both “taking the blame” and “making choices.” The former refers to acknowledging the perspectives of health care professionals and assuming the blame for recent experiences of symptoms. Behaviors identified as risk factors by nurses and physicians (e.g., smoking, an excessive use of alcohol, consuming a poor-quality diet, or a sedentary lifestyle) were verbally acknowledged by the participants to be part of the reason that a cardiac event had occurred. Making choices refers to women’s efforts to incorporate the advice of health care professionals, in relation to lifestyle changes. In each case, if a woman was diagnosed with a cardiac event, she sometimes made drastic alterations in her daily habits. For women who did not receive a cardiac diagnosis, taking the blame and making choices were equally prominent, but in different ways. In integrating knowledge and experiences, the women made sense of new information and used that information to make choices about the future. During this stage, the women synthesized three sources of information to make sense of recent events as a whole, and then integrated the understanding gained into the context of their lives. First, the women had a tentative diagnosis upon their discharge from the hospital (e.g., cardiac or noncardiac symptoms). As well, the women drew on information about their individual risk behaviors in relation to the development of heart disease (e.g., in relation to alcohol consumption, exercise, diet), gleaned from interactions with

health care professionals during the ED visit. Finally, the women drew on their experiences of care in the ED to position recent events within the context of their lives and to synthesize all of the knowledge and experience they had gained in order that they might make decisions in future situations. In this phase, the women persevered in their efforts to preserve their images of themselves as healthy (e.g., making decisions to change risk behaviors) and socially worthy (e.g., not wasting health care resources). In this phase, the women struggled to reconcile their individual perspectives and values with those of their health care providers. Reconciling new information with their images of themselves as healthy and capable was sometimes challenging, particularly if a woman had received a cardiac diagnosis. Nevertheless, the women were able to maintain personal, social, and physical integrity by balancing their new diagnoses with a commitment to changing health behaviors with regard to risk factors. Conversely, if a noncardiac diagnosis was received, the women in this study reported no intention to alter their present or future health behaviors. Whether or not the participants in this study received a diagnosis of a cardiac illness, they were well aware of the judgments of health care professionals, as has been previously discussed. However, the women who were diagnosed as having had a cardiac event accepted judgments about the appropriateness their health behaviors, whether consciously or unconsciously. For example, one woman, speaking about her decision to stop smoking completely and abruptly after 44 years, stated, But it’s my own decision. Then again, maybe he [the general practitioner] happened to mention, “Did I smoke,” and I said “Yeah”; maybe somewhere in my . . . in the back of my mind, I’m thinking, “Hey, this is a good warning . . . time to wake up and smell the roses.”

Similarly, another participant talked about her hope that exercise, dietary changes, and refraining from alcohol might balance the effects of smoking over the past decades. “I’ve never [had any risk factors] but the smoking . . . so that’s got to be in my favor.” Being diagnosed with a cardiac event resulted in a re-evaluation of personal risk and risk behaviors. Fear was a motivation for behavior change as one of the women stated, “That scared me. I don’t want to be in that place again.” As well, the reactions of family members served as a motivation for behavior change in relation to health behaviors. As one participant

1472 Qualitative Health Research

exclaimed, “All my family . . . they came out of the woodwork. You don’t see them for years and all of a sudden, ‘Jesus! I am dying!’” Accordingly, the women in this study who received a cardiac diagnosis took aggressive action. In the first example, a participant drastically altered her diet to include food high in fiber and low in fat. I’ve been trying to look at the labels to get checkmark things that say healthy which means hearty hearts or whatever and cereal I’ve tried to change sort of gradually. I bought some like Kashi cereal instead of just Shreddies and Rice Krispies and stuff. I use only Splenda. I haven’t used sugar; I don’t smoke [anymore], I don’t drink coffee, I don’t drink alcohol so I figure all those things are probably in my favor as not necessarily just to do with food. Exercise too . . . that was the other thing. I started out mostly once a day, but only 5 or 6 blocks, and lately I haven’t done that much. I do like to swim but my pool is closed down until next Monday so I’m going to start back on that. But I think all things considered I’ve taken a lot of . . . when I look at the whole picture I’ve probably done a lot to adjust [after my heart attack].

Another participant started exercising as a result of the instructions of her physician. “The doctor told me when I go home to walk like for 10 minutes a day for the first week or so.” And, after being diagnosed with a heart attack, she stopped drinking alcohol, which had been a health issue for her in the past. “I haven’t had anything. Absolutely, I don’t want nothing! I don’t want any alcohol, I don’t want nothing. The only thing I’m having a hard time with is quitting smoking. That’s tough.” Yet another woman started attending a Healthy Heart program on a weekly basis after receiving a cardiac diagnosis. Although prior to having a heart attack the women had information about the risks associated with smoking, for example, they did not take action. In contrast, after being diagnosed, the women in this study adopted that information and used their knowledge to alter their lifestyles, illustrating the power of interactions with health care professionals in the context of an acute cardiac event. In essence, this group of women had been given a new piece of information— a cardiac diagnosis—and they acted on this knowledge, undertaking changes to their lifestyles. In this case, new knowledge was personalized and this resulted in the uptake of information and, ultimately, behavior change. The responses of the women who did not receive a cardiac diagnosis differed substantially from those

who did. For women without a cardiac diagnosis, making sense of the problem also involved a reflection on the events that had occurred. However, far from planning to alter their lifestyles in relation to risk behaviors, these women chose to maintain the status quo. For example, without exception, when asked what they would do if their symptoms recurred, this group of women stated that their behavior would be unchanged. The participants who “watched and waited” said that they would choose the same course of action (or inaction) if their symptoms recurred. Similarly, if a participant sought treatment immediately, she reported that she would do so again. The women without a cardiac diagnosis had no new information on which to draw. Their experiences of ED care appeared to make no difference to their intention to return. Instead, social concerns about embarrassment and wasting resources, as well as prior knowledge about the ED, were the major factors influencing decisions about treatment seeking. For these women taking the blame involved self-castigation for “wasting people’s time” and acknowledging feelings of intense embarrassment about the “false alarm.” Making choices was about expressing the intention to choose a similar course of action for future events.

Discussion Gender and TSD, in the context of symptoms of cardiac illness, remains a clinical issue given the time-dependent nature of available therapies. The results of this study, investigating prehospital decision making and experiences of ED care, illustrate that women resisted any disruption to their daily lives and this influenced their decisions about treatment seeking. Family responsibilities, employment roles, and social roles were generally judged to be more of a priority than an experience of cardiac symptoms. Framing TSD as an individual phenomenon might be limiting our ability to address the problem, as this approach problematizes individual behavior and ignores both the social context of health behavior and the inadequacy of system responses. Framing TSD as a system phenomenon, and understanding TSD as related to the women’s efforts to maintain personal, social, and physical integrity, might result in interventions that will address the issue of TSD comprehensively, as we have tried unsuccessfully to do for the past 25 years. The results of this study hint at the possible benefits of reconceptualizing TSD as more than an issue of individual decision making. Such

Turris, Johnson / Maintaining Integrity 1473

development might support researchers in moving away from the exploration of purely individual factors and toward examining the spectrum of care and how that care might be influencing women’s decisions about treatment seeking. This study adds to the literature on TSD by focusing attention primarily on women who do not yet have a cardiac diagnosis. Understanding how decisions about treatment seeking evolve, before cardiac events occur, is a proactive approach that might allow us to generate more effective interventions. Schoenberg, Peters, and Drew (2003) studied the experiences of 40 women diagnosed with heart disease, or with conditions that put them at risk for the development of heart disease. In common with this study, they focused not on women who had a heart attack, but on women who might have one. They identified four themes: symptom uncertainty, problematic interactions with health care providers, competing social demands, and structural barriers to medical care. The third theme is in keeping with the findings of this project—competing work, family, or social responsibilities took precedence. Similarly, Dempsey, Dracup, and Moser (1995) studied women’s prehospital experiences using a GT approach (and the self-regulation model of illness behavior). They interviewed 16 women following a heart attack. They reported two core categories: maintaining control and relinquishing control. The women reported that “commitments” were often an overriding concern, as was concern for others. They made a trio of recommendations: teach that women and men are at risk; talk to women about the fact that they will want to minimize symptoms; and provide information about options so women gain control, rather than feeling that they are giving up control by seeking advice. The role of stress discourse in women’s experiences of their cardiac symptoms was a noteworthy finding of this study. Without exception, participants raised the effects of stress (and an implied failure to cope) as a partial or complete explanation for their experience of symptoms. In essence, the women in this study berated themselves for a perceived inability to cope with stress, which they speculated might be the root cause of their symptoms. Stress was not viewed as a legitimate explanation for symptoms, but rather a cause for self-castigation. A perceived failure to cope was viewed by the participants as a moral failure. Stress, as an explanation for symptoms, on some level reassured the women that in the future, if

they managed to cope more effectively, symptoms would not recur. On the other hand, this meant that the sole responsibility for managing stress lay with the individual woman. Notable was the absence of a consideration of social and organizational factors that might create milieus in which stress is embedded. This is consistent with recent findings by researchers that women view stress primarily as an individual experience rather than examining the culture of a particular place as part of the explanation (Harkness et al., 2005). Although several of the women touched on social explanations for stress, such as the work environment, they expressed a perception that notwithstanding these challenges, they should (as individuals) be able to cope (e.g., manage stress effectively and so not experience physical illness as a manifestation of that stress). This pervasive discourse exerted a typically negative influence on treatmentseeking decisions. Foucault (1980) and feminist writers view the body as a site of power and view discourse as a way of producing and sustaining power. In this study, multiple discourses shaped women’s efforts to maintain integrity, and influenced their subsequent experiences of ED care. The discourse of resource scarcity was apparent in the statements of the women who were reluctant to seek treatment for fear of “bothering” health care professionals, perceived to be both busy and overworked. The discourse of acuity was obvious in the belief, held by the women, that an illness or injury warranting a visit to the ED should be visible to any bystander, a belief that often functioned to delay treatment seeking. The discourse of deservedness was apparent in the common statement that others were more in need of care and in the “mea culpa” position adopted by women who had risk factors associated with the development of heart disease such as smoking or a lack of exercise. The women in this study also experienced the influence of biomedical discourse. As identified by Malone (1998), in the ED “a kind of legitimacy thus inheres to conditions that can be rapidly and definitively treated” (p. 810). In the ED, both the physical environment and nursing practice were shaped by biomedical discourse in ways that were at times unfriendly for patients. As Heslop (1998) argued, ED nursing care takes place within a biomedical discourse that dominates the practice of nursing and determines acceptable practices. The application of professional knowledge to the bodies of the women has undoubtedly positive effects on the health and wellness of the women. However,

1474 Qualitative Health Research

when this knowledge was extended and applied to judgments about deservedness, the impact was not positive. For example, risk avoidance as a moral exercise was a theme in this study; women were well aware that health care professionals judged the actions of the participants in relation to risk behaviors. The ED is a place where the private becomes public. An obligation to remain healthy is a moral dimension of citizenship; health and health care are not neutral or apolitical. In summary, although several studies have explored women living with coronary heart disease and identified women’s gender-specific experiences (e.g., Kristofferson, Lofmark, & Carlsson, 2003; Lacharity, 1999; Rosenfeld, Lindauer, & Darney, 2005; Sutherland & Jensen, 2000), few studies have examined women’s experiences of seeking help for the potential symptoms of an acute cardiac event. This study presents new and unique information that could potentially address a significant gap in research knowledge.

Limitations When interpreting the results of this study, there are several limitations to be considered. With regard to the sample, at least two groups of women were not represented: women who were critically ill and women who did not present for treatment in an ED. Only women who presented to an ED for treatment of symptoms of potential cardiac disease were recruited. The accounts and experiences of women who did not come to an ED for treatment are missing. It is possible, and indeed probable, that decision making differs for this group in comparison to the sample for this study. In addition, we were unable to recruit critically ill women for participation in this study (because of hemodynamic instability). The inclusion of such women in the sample might have altered the theory generated. Insider status might have contributed to the introduction of bias during both data collection and analysis, although actions were undertaken to reveal and minimize the effect of insider status. Finally, it is important to bear in mind that this is a single study. As such, the results add to our understanding about treatment-seeking delay and experiences of ED care, but are an insufficient basis on which to recommend policy or practice changes.

Conclusions For the participants, efforts to maintain personal, social, and physical integrity were to a great extent

situated. That is, maintaining integrity was both enabled and constrained by the place in which events unfolded. In home and work environments the women were free to evaluate their situation and to take independent action to address their experience of symptoms. In contrast, in the ED, the women were attempting to maintain personal, social, and physical integrity within a structure that limited their ability to act and shaped the meaning they ascribed to their experiences.

Note 1. In vivo codes arise directly from the data and are often labeled in terms used by the study participants.

References American Heart Association. (2004). Heart disease and stroke statistics—2004 update. Dallas, TX: Author. American Heart Association Statistics Committee. (2007). Heart disease and stroke statistics, 2007 update. Circulation, 115, 1-105. Antman, E. M., Anbe, D. T., Armstrong, P. W., Bates, E. R., Green, L. A., Hand, M., et al. (2004). ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: Executive summary. Circulation, 110, 588-636. Ashton, K. C. (1999). How men and women with heart disease seek care: The delay experience. Progress in Cardiovascular Nursing, 14, 53-60. Banks, A. D., & Dracup, K. (2006). Factors associated with prolonged prehospital delay of African Americans with acute myocardial infarction. American Journal of Critical Care, 15, 149-157. Banks, A. D., & Dracup, K. (2007). Are there gender differences in the reasons why African Americans delay in seeking medical help for symptoms of an acute myocardial infarction? Ethnicity and Disease, 17(2), 221-227. Berger, J. S., Sanborn, T. A., Sherman, W., & Brown, D. L. (2006). Influence of sex on in-hospital outcomes and longterm survival after contemporary percutaneous coronary intervention. American Heart Journal, 151, 1026-1031. Bierman, A. S. (2007). Sex matters: Gender disparities in quality and outcomes of care. Canadian Medical Association Journal, 177(12), 1520-1521. Blomkalns, A. L., Chen, A. Y., Hochman, J. S., Peterson, E. D., Trynosky, K., Diercks, D. B., et al. (2005). Gender disparities in the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes. Journal of the American College of Cardiology, 45, 832-837. Bowker, T. J., Turner, R. M., Wood, T. L., Roberts, T. L., Curzen, M., Gandhi, M., et al. (2000). A national survey of acute myocardial infarction and ischaemia (SAMII) in the UK: Characteristics, management and in-hospital outcome in women compared to men in patients under 70 years. European Heart Journal, 21, 1458-1463. Bunde, J., & Martin, R. (2006). Depression and prehospital delay in the context of myocardial infarction. Psychosomatic Medicine, 68, 51-57. Canadian Heart and Stroke Foundation. (2008). Warning Signals. Retrieved August 21, 2008, from http://www.heartandstroke

Turris, Johnson / Maintaining Integrity 1475

.com/site/c.iklQLcMWJtE/b.3483917/k.BF6E/Heart_attack_ warning_signals.htm Canadian Institute of Health Information. (2006). Health care in Canada. Ottawa: Author. Charmaz, K., & Mitchell, R. G. (2000). Grounded theory in ethnography. In P. Atkinson, A. Coffey, S. Delamont, J. Lofland, & L. Lofland (Eds.), Handbook of ethnography (pp. 160-174). London: Sage. Cornado, B. E., Pope, H. J., Griffith, J. L., Beshansky, J. R., & Selder, H. P. (2004). Clinical features, triage, and outcome of patients presenting to the ED with suspected acute coronary syndromes but without pain: A multicenter study. American Journal of Emergency Medicine, 22, 568-574. Cox, J. L., Bata, I. R., Gregor, R. D., Johnstone, D. E., & Wolf, H. K. (2006). Trend in event rate and case fatality of patients hospitalized with myocardial infarction between 1984 and 2001. Canadian Journal of Physiology and Pharmacology, 844, 121-127. Dempsey, S. J., Dracup, K., & Moser, D. K. (1995). Women’s decision to seek care for symptoms of acute myocardial infarction. Heart and Lung, 24, 444-456. Dracup, K., Moser, D. K., McKinley, S. M., Ball, C., Yamasaki, K., Kim, C. J., et al. (2003). An international perspective on the time to treatment for acute myocardial infarction. Journal of Nursing Scholarship, 34, 317-323. Finnegan, J. R., Meischke, H., Zapka, J. G., Leviton, L., Meshack, A., Benjamin-Garner, R., et al. (2000). Patient delay in seeking care for heart attack symptoms: Findings from focus groups conducted in five U.S. regions. Preventive Medicine, 31, 205-213. Foucault, M. (1980). Power/knowledge. Selected interviews and other writings 1972-1977. Colin Gordon (Ed.). New York: Pantheon Books. Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto. (1995). Epidemiology of avoidable delay in the care of patients with acute myocardial infarction in Italy: A GISSI-generated study. Archives of Internal Medicine, 155, 1481-1488. Glaser, B. G. (1978). Theoretical sensitivity: Advances in the methodology of grounded theory. Mill Valley, CA: Sociology Press. Glaser, B. G. (1992). Basics of grounded theory. Mill Valley, CA: Sociology Press. Glaser, B. G. (1998). Doing grounded theory: Issues and discussion. Mill Valley, CA: Sociology Press. Glaser, B. G., & Strauss, A. A. (1999). The discovery of grounded theory: Strategies for qualitative research (2nd ed.). New York: Aldine de Gruyter. Goldberg, R. J., Gurwitz, J. H., & Gore, J. M. (1999). Duration of, and temporal trends (1994-1997) in, prehospital delay in patients with acute myocardial infarction. Archives of Internal Medicine, 159, 2141-2147. Goldberg, R. J., Steg, P. G., Sadiq, I., Granger, C. B., Jackson, E. A., Budaj, A., et al. (2002). Extent of, and factors associated with, delay to hospital presentation in patients with acute coronary disease (the GRACE registry). American Journal of Cardiology, 89, 791-796. Hammond, J., Salamonson, Y., Davidson, P., Everett, B., & Andrew, S. (2007). Why do women underestimate the risk of cardiac disease? Australian Critical Care, 20(2), 53-59. Harkness, A. M. B., Long, B. C., Bermbach, N., Patterson, K., Jordan, S., & Kahn, H. (2005). Talking about work stress:

Discourse analysis and implications for stress interventions. Work & Stress, 19, 121-136. Heslop, L. (1998). A discursive exploration of nursing work in the hospital emergency setting. Nursing Inquiry, 5, 87-95. Ho, M. T., Eisenberg, M. S., Litwin, P. E., Schaeffer, S. M., & Damon, S. K. (1989). Delay between onset of chest pain and seeking medical care: The effect of public education. Annals of Emergency Medicine, 18, 727-731. Kristofferson, M., Lofmark, R., & Carlsson, M. (2003). Myocardial infarction: Gender differences in coping and social support. Journal of Advanced Nursing, 44, 360-374. Lacharity, L. A. (1999). The experiences of younger women with coronary artery disease. Journal of Women’s Health and Gender-Based Medicine, 8(6), 773-785. Lansky, A. J., Cody, P., Costa, R. A., Tsuchiya, Y., Brodie, B. R., Cox, D. A., et al. (2005). Gender differences in outcomes after primary angioplasty versus primary stenting with and without abciximab for acute myocardial infarction: Results of the controlled abciximab and device investigation to lower late angioplasty complications (CADILLAC) trial. Circulation, 111, 1611-1618. Lefler, L. L., & Bondy, K. N. (2004). Women’s delay in seeking treatment with myocardial infarction: A meta-synthesis. Journal of Cardiovascular Nursing, 19, 251-268. Leitch, J. W., Birbara, T., Freedman, B., Wilcox, I., & Harris, P. J. (1989). Factors influencing the time from onset of chest pain to arrival at hospital. Medical Journal of Australia, 150, 6-8. Lesneski, L., & Morton, P. (2000). Delay in seeking treatment for acute myocardial infarction: Why? Journal of Emergency Nursing, 26, 125-129. Lockyer, L. (2002). The construction of a modern epidemic: The implications for women of the gendering of coronary heart disease. Journal of Advanced Nursing, 39, 432-440. Malone, R. E. (1998). Whither the almshouse? Overutilization and the role of the emergency department. Journal of Health Politics, Policy and Law, 23, 795-832. McKinley, S. M., Dracup, K., Moser, D. K., Ball, C., Yamasaki, K., Kim, C.-J., et al. (2004). International comparison of factors associated with delay in presentations for AMI treatment. European Journal of Cardiovascular Nursing, 3, 225-230. Moser, D. K., & Dracup, K. (1993). Gender differences in treatment-seeking delay in acute myocardial infarction. Progress in Cardiovascular Nursing, 8, 6-12. Moser, D. K., McKinley, S. M., Dracup, K., & Chung, M. L. (2005). Gender difference in reasons patients delay in seeking treatment for acute myocardial infarction symptoms. Patient Education and Counselling, 56, 45-54. O’Brian, J., & Fothergill-Bourbonnais, F. (2004). The experience of trauma resuscitation in the emergency department: Themes from seven patients. Journal of Emergency Nursing, 30, 216-224. O’Donnell, S., Condell, S., Begley, C., & Fitzgerald, T. (2005). In-hospital care pathway delays: Gender and myocardial infarction. Journal of Advanced Nursing, 52, 14-21. Pattenden, J., Watt, I., Lewin, R. J. P., & Stanford, N. (2002). Decision making processes in people with symptoms of acute myocardial infarction: Qualitative study. British Medical Journal, 324, 1006-1009. Pattendon, J., Watt, I., Lewin, R. J. P., & Stanford, N. (2003). Decision making processes in people with symptoms of acute

1476 Qualitative Health Research

myocardial infarction: Qualitative study. British Medical Journal, 324(7344), 1006-1009. Polit, D. F., & Hungler, B. P. (1999). The analysis of qualitative data. In D. F. Polit & B. P. Hungler (Eds.), Nursing research: Principles and methods (6th ed., pp. 573-590). Philadelphia: Lippincott. Redley, B., Beanland, C., & Botti, M. (2003). Accompanying critically ill relatives in emergency departments. Journal of Advanced Nursing, 44, 88-98. Reimer, W. J. M., Gitt, A. K., Borersma, E., & Simmons, M. L. (2006). Euro heart survey: Cardiovascular diseases in Europe 2006. Nice, France: European Heart Society. Richards, H. M., Reid, M. E., & Watt, G. C. M. (2002). Why do men and women respond differently to chest pain? Journal of the American Medical Women’s Association, 57(2), 79-81. Rogers, M. A. M., Langa, K. M., Kim, C. J., Nallamothu, B. K., McMahon, L. F., & Malani, P. N., et al. (2006). Contribution of infection to increased mortality in women after cardiac surgery. Archive of Internal Medicine, 166, 437-443. Rosenfeld, A. G., Lindauer, A., & Darney, B. G. (2005). Understanding treatment seeking delay in women with acute myocardial infarction: Descriptions of decision-making patterns. American Journal of Critical Care, 14, 285-293. Schoenberg, N. E., Peters, J. C., & Drew, E. M. (2003). Unraveling the mysteries of timing: Women’s perceptions about time to treatment for cardiac symptoms. Social Science and Medicine, 56, 271-284. Statistics Canada. (2004). Age-standardized mortality rates (female). Retrieved October 26, 2004, from http://www.40 .statcan.ca/101/cst01/health30a.htm Sutherland, B., & Jensen, L. (2000). Living with change: Elderly women’s perceptions of having a myocardial infarction Qualitative Health Research, 10, 661-676.

Walsh, M. (1993). Delaying attendance at A & E departments. Nursing Standard, 7(24), 33-35. White, A., & Lockyer, L. (2001). Tackling coronary heart disease: A gender sensitive approach is needed. British Medical Journal, 323(7320), 1016-1017. White, A. K., & Johnson, M. (2000). Men making sense of their chest pain—Niggles, doubts, and denials. Journal of Clinical Nursing, 9, 534-541. Woodfield, S. L., Lundergan, C. F., Feiner, J., Thompson, M. A., Rohrbeck, S. C., Deychak, Y., et al. (1997). Gender and acute myocardial infarction: Is there a different response to thrombolysis? Journal of the American College of Cardiology, 29(1), 35-42. Wu, Y., Zhang, Y., Li, Y., Hong, B., & Huang, C. (2004). Factors associated with the extent of care-seeking delay for patients with acute myocardial infarction in Beijing. Chinese Medical Journal, 117, 1772-1777. Zerwic, J. J. (1999). Patient delay in seeking treatment for acute myocardial infarction symptoms. Journal of Cardiovascular Nursing, 13, 21-32. Zerwic, J. J., Ryan, C. J., DeVon, H. A., & Drell, M. J. (2003). Treatment seeking for acute myocardial infarction symptoms. Nursing Research, 52, 159-167. Sheila A. Turris, RN, PhD, is a faculty member in the Emergency Nursing Program at the British Columbia Institute of Technology in Burnaby, British Columbia, Canada. Joy L. Johnson, PhD, RN, FCAHS, is a professor at the School of Nursing, University of British Columbia, in Vancouver, British Columbia, Canada, and the scientific director of the Institute of Gender and Health, Canadian Institutes of Health Research, in Ottawa, Ontario, Canada.