MIMI

        

WW‘lfll‘llllfillflllfilmH\l|‘.\‘WllH‘|H\

(0CD

N
Iw
m

    

Date

0-7639

                  

:I

                 

 

   

 

Ulillllllllllllllllllllllll

23 01409 8614

LIBRARY
Michigan State
University

 

 

 

 

This is to certify that the

thesis entitled

CHEST PAIN QUALITY DESCRIPTION AS A
PREDICI'OR OF ACUTE MYOCARDIAL ISCI-IEMIA:
IS THERE A GENDER DIFFERENCE?

presented by

Barbara Clare H. Jaquith

has been accepted towards fulfillment
of the requirements for

IVIaSter Of SCieIlCe degree in NurSing

 

College of Nursing

/ Z/SZVZ

/Major professor

1/ 27/75-

MS U is an Affirmative Action/Equal Opportunity Institution

' “ ‘éfl'flq ’M‘.‘_#"‘— — —

 

 

PLACE N RETURN BOX to remove title checkout from your record.
To AVOID FINES return on or More dete due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU IeAn Affirmative Action/Bond Opportunity lnetituion
W

m1

CHEST PAIN QUALITY DESCRIPTION AS A
PREDICTOR OF ACUTE MYOCARDIAL ISCHEMIA:
IS THERE A GENDER DIFFERENCE?

By

Barbara Clare H. Jaquith

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE IN NURSING

College of Nursing

1 995

ABSTRACT

CHEST PAIN QUALITY DESCRIPTION AS A
PREDICTOR OF ACUTE MYOCARDIAL ISCHEMIA:
IS THERE A GENDER DIFFERENCE?

By

Barbara Clare H. Jaquith

Chest pain is a deciding factor leading patients with cardiac-type symptoms to seek health
care. Subjective description of chest pain quality may be the first definitive information the
health care provider has with which to assess patients initially, and base clinical decisions
regarding the risk for acute myocardial ischemia.

A retrospective chart review of emergent care investigated three gender-based research
questions: 1. are there differences in how women and men describe chest pain quality; 2. do
these descriptions predict hospital admission; and 3. are descriptive adjectives diagnostically
accurate predictors of elevated serum creatine kinase-MB results, and indicative of myocardial
ischemia.

Results indicated a significant difference in chest pain quality description by women and
men. This difference may cue the possibility of admission, but not the probability of elevated
initial CKMB results indicative of cardiac disease severity. Advanced nursing practice
implications related to assessment and clinical decision-making in the primary care setting are

discussed.

Copyright by
BARBARA CLARE H. JAQUITH
1995

With love to my husband Stephen,
and to my children Anne, and Johnathan.

"God rarely allows a soul to see how great a blessing he is."

-Oswald Chambers

ACKNOWLEDGMENTS

I am grateful for the valuable support and assistance I have received in the completion of
this research. Manfred Stommel, chairperson of my thesis committee, provided expert guidance
through the research process, and ongoing encouragement. Patricia Peek, and Catherine Lein
have guided content and nursing theory portions of this project with helpful suggestions, draft
reviews, and cheerful support.

Several people at Northern Michigan Hospital provided assistance, namely, the Medical
Records clerks at Northern Michigan Hospital who helped with chart review; and my nursing and
physician colleagues who cared for the patients with chest pain, documented patient chest pain
descriptions, and kept up a running curiosity and inquiry as to the progress of research and
writing.

I have a special thanks to my fellow students Diana Hayes, Susan Jones, and Tara Conti
for their interest and the empathy they provided as I completed this research.

Finally, and mostly, i am grateful to and thankful for my family, who have offered their love,

listened patiently, carried on, and missed me when I spent too much time at the computer.

TABLE OF CONTENTS

LIST OF TABLES ...................................................................................... viii
LIST OF FIGURES .................................................................................... ix
INTRODUCTION ....................................................................................... 1
Definition of Problem ........................................................................... 1
Benner and Clinical Decision-Making ................................................. 2
Research Problem Statement ............................................................. 3
Conceptual Definition of Variables ...................................................... 4
Gender ................................................................................................ 4
Acute Myocardial lschemia ................................................................ 4
Chest Pain Quality ............................................................................... 6
CKMB .................................................................................................. 6
Other Variables .................................................................................... 7
LITERATURE REVIEW .............................................................................. 7
Sociodemographics of CHD ................................................................ 7
Gender Differences in Treatment for AMI ........................................... 1O
Perception of Illness, and Care-Seeking Behaviors ............................ 12
Clinical Decision-Making ..................................................................... 13
CKMB as a Diagnostic Tool ................................................................. 15
METHODS ................................................................................................. 18
Research Design ................................................................................. 18
Sample ................................................................................................. 19
Data Collection and Recording ........................................................... 20
Ethics ................................................................................................... 20
Operational Definition of Variables ...................................................... 20
Data Analysis Procedures .................................................................... 21
RESULTS ................................................................................................... 22
Demographics ...................................................................................... 22
Research Questions ............................................................................ 27
Other Findings .................................................................................... 35
Interpretation of Findings ..................................................................... 35
DISCUSSION ............................................................................................. 37
Implications for Advanced Nursing Practice and Primary Care ........... 37
Implications for Current Literature and Future Research .................... 39
Limitations of Research ...................................................................... 41
Summary .............................................................................................. 43

vi

Table of Contents (Continued):

LIST OF APPENDICES
A Study Approval Letters ................................................................... 44
B Chemistry Department Technical Manual- CKMB .......................... 47
REFERENCES ........................................................................................... 48

vii

LIST OF TABLES

Table 1: Admission Statistics for Diagnosis and Age by Gender .............. 25
Table 2: Comparison of Admitting Diagnosis and

Discharge Diagnosis by Gender ................................................ 26
Table 3: Descriptive Adjectives ................................................................ 28
Table 4: Adjective Use by Gender ........................................................... 29
Table 5: Prediction of Hospital Admission to a Critical Care Area ............ 31
Table 6a: Overall Prediction of Elevated CK/CKMB Results .................... 33

Table 6b: Gender-Based Predicted Elevation of Peak CKMB Results ..... 34

viii

LIST OF FIGURES

Figure 1: Patterns and timing of elevation of creatine kinase (CK),
creatine kinase-MB (CKMB), and other cardiac enzymes ......... 15

INTRODUCTION

Chest pain, as a manifestation of coronary heart disease (CHD), is often the deciding
factor causing patients to seek care for cardiac health problems (Schroeder, Lamb 8. Hu, 1978).
Patients who seek professional help usually come to the primary care provider, or the hospital
Emergency Department (ED). A patient's description of the quality of their chest pain may be the
first definitive information the health care provider has on which to base triage decisions.

Epstein (1990) explores the creation of clinical prediction rules as a way of guiding usage
of diagnostic tests to improve patient care. The challenge of improving diagnostic accuracy in
the present era of increasing use of high technology may fall to the primary care provider, be
they the physician, the Advanced Practice Nurse (APN), or the staff nurse involved in patient
assessment and triage. The role of the health care provider requires accurate and timely
assessment of initial patient complaints on presentation to the clinic or ED, and makes
necessary the early recognition of patients at high risk for AMI. This allows speedy access to
appropriate pharmacological (thromboiytic) or mechanical (surgical) treatments that can aid in
preservation or salvage of cardiac muscle.

Evaluation of chest pain is complicated by the lack of universally occurring historic,
clinical, and laboratory results indicative of acute myocardial ischemia (AMI) (Hedges, Rouan,
Toltzis, Goldstein-Wayne, & Stein, 1987). While those with acute chest pain tend to have
certain characteristics that are more common in people with infarction- male gender, increased
age, smoking history, hypertension, obesity, diabetes- none are sufficiently discriminatory to be
useful for diagnosis.

Nurses often are confronted with medical emergencies such as chest pain, which requires

immediate attention. This recognition component of assessment (chest pain quality) has

increased dramatically over the last two decades. Nurses are required to make graded
qualitative distinctions, and initiate care based on clinical decision-making, many times before
objective diagnostic results may be available.

Benner (1984) addressed these issues involving nursing and clinical decision making in
her studies of how nurses achieve excellence and power in clinical practice through the
application of the Dreyfus model of skill acquisition. She recognized that there are differences
between practical knowledge/skills— "know how", and theoretical knowledge- "know that".
Benner believed that theoretical knowledge is embedded in nursing practice in the form of
clinical knowledge accrued over time. The establishment of interactional relationships between
events and practical knowledge shapes questions for theory development and enhances
predictability for similar events.

Benner gathered that a nurse develops expertise when she/he tests and refines
propositions and hypotheses in actual practice situations through challenges to preconceived
notions. Experience is a prerequisite for expertise in nursing practice, and is reflective of
differing expositional and problem solving techniques between novice and expert nurses,
requiring the use of past situations as paradigms. As nurses compare their assessments in
actual situations, such as caring for a patient with chest pain, graded qualitative distinctions can
emerge from expert nursing practice and be shared as clinical judgment or decision making.

Benner realized that nurses are faced with two conflicting mandates: to individualize
patient care, and to minimize errors through use of standards of patient care. The intent of this
research is to strengthen assessment and history-taking as a tool used in triage and clinical
decision-making for patients presenting with chest pain. Knowledge in expert nursing practice
transcends norms and standardization. Searching for clinical relationships between history and
definitive laboratory values may fill gaps in knowledge related to assessment skills for those in
primary and emergency care, and will strengthen nursings‘ role as a communicator and

diagnostician.

Research Problem Statement

CHD strikes men more frequently, beginning in the middle years, and is usually perceived
by the public as a male disease. Despite its greater prevalence among men, coronary heart
disease is also a major threat to women's health, and accounts for 250,000 deaths in women of
all ages in the United States each year. The incidence among women has risen steadily since
1950, while it has declined among men in the same period (Ayanian & Epstein, 1991; Eysmann
& Douglas, 1992). For women, the death rate from CHD is three times higher than from breast
and lung cancer combined.

The proposed study evaluates three research questions. These questions are based on
two of the three diagnostic criteria for Ml from the World Health Organization (WHO, 1959), i.e.,
clinical history, and elevated cardiac enzyme levels. The third criterion, diagnostic
electrocardiography (ECG) changes, will not be used in the present study. Gibler et al. (1990)
noted that diagnostic 12-Iead ECG changes are absent in up to 50% of the patients presenting to
the ED with AMI. New techniques using 18-lead ECGs show improved diagnostic capabilities,
but this procedure is not yet standard practice at the study site.

The first question is addressed to the possibility of systematic differences in how women
and men describe their primary symptom of chest pain on presentation to the ED. Bickell et al.
(1992) suggest that "perhaps men and women perceive and express symptoms (of chest pain)
differently; ...or that physicians hear and interpret these expressions differentljr' (p. 796). The
seriousness with which women view the symptoms of angina, and the language they use to
describe their symptoms, may well influence the care they receive. Perhaps women use a
vocabulary different from that of men to describe similar types of perceived chest pain, reflecting
differences in life experiences, socialization, or actual experiences of pain. An example is a man
who may use terms as "crushing," or "kicked in the chest" to describe chest pain, whereas a
woman may be less likely to use these terms, or to use others in describing similar pain.

The second question asks whether womens' and mens' descriptions of chest pain quality

predict hospital admission to a critical care area and, if so, do the same words predict admission

for women and for men? Correlations between adjectives employed by patients and decisions
to admit could indicate that physicians stratify patients into high-risk (requiring hospitalization),
or low-risk (safe to discharge), in part, based on their verbalizations.

The third question analyzes whether certain descriptive adjectives used by women or men
to describe chest pain qualities are diagnostically accurate predictors of individually elevated
serum creatine kinase—MB (CKMB) results, and therefore indicative of AMI. Evidence of such
correlations may be gender specific, and may increase diagnostic accuracy, aiding in risk
stratification regarding choice of treatment, and need for hospital admission (Plotnick & Fisher,
1985). Compilation of a list of "high-risk" adjectives could lead to training of physicians and
nurses in the identification of high-risk patients during initial assessments, and lead to more
timely and efficient care for potential cardiac patients. This researcher recognizes that the
descriptive adjectives used to predict admission may not be the same words that indicate
elevated CKMB results.

Conceptui Definition of Variables

Gender

Webster defines gender as "a person‘s seX' (Neufeldt & Sparks, 1990, p. 246). This
researcher accepts this definition for this variable.

Acute Myocardial lschemia

AMI includes new—onset angina pectoris, unstable Dre—existing angina pectoris, and acute
myocardial infarction (Pozen et al., 1980). The mechanism of myocardial ischemia helps to
differentiate among these three categorizations of AMI. Myocardial ischemia occurs when the
myocardial demand for oxygen exceeds oxygen supply and metabolite removal, most often due
to vasospasm, or vascular occlusion. This results in ineffective cell functions. lschemic
myocardial muscle undergoes anaerobic metabolism. This leads to release of lactic acid,
regional ischemia, and impaired left ventricular function (Bullock & Rosendahl, 1984). lschemia
is considered reversible, and restoration of cellular function can occur with restoration of oxygen

to the affected muscle.

Angina pectoris is a manifestation of myocardial ischemia, characterized by squeezing,
substemal chest pain that is often described as a feeling of tightness or fullness. Physiological
response to the pain includes pallor, perspiration, and dyspnea. Angina may be precipitated by
physical or emotional stress, and relieved by rest. ECG changes may include T-wave inversion,
and S-T segment depression. Laboratory findings are usually normal (Bullock & Rosendahl,

1 984).

Pozen et al. (1980) define those with new-onset angina as patients presenting with
symptoms of angina pectoris for the first time, subsequently documented by S-T wave changes
with pain, a positive exercise stress test, or a positive coronaryarteriogram. These patients are
considered "stable".

These authors go on to describe someone with unstable, pre-existing angina as one with a
previously stable history of angina lasting at least three months, whose frequencies of episodes
of angina have increased, or who had a worsening clinical picture. Bullock and Rosendahl
(1984) refer to this pattern as preinfarction angina, indicative of CHD progression. Schroeder,
Lamb, and Hu (1978) relate unstable angina to the presence of prodromal symptoms, but
suggest the presence of these symptoms is not necessarily predictive of myocardial infarction.

Myocardial infarction (MI) results from prolonged ischemia to the myocardium, causing
irreversible cell damage and cell death. Coronary artery flow is generally impeded by a
combination of vasospasm, platelet aggregation, and thrombus formation. Myocardial tissue
death results in increased demand for cardiac output from the surviving myocardium,
dysrhythmias (Hedges & Kobemick, 1988); varied clinical manifestations; and in the release of
intracellular enzymes, including creatine kinase (Bullock & Rosendahl, 1984).

This researcher accepts the categorization of Pozen et al. for acute myocardial ischemic
disease for this study. Stable angina will be defined as new-onset anginal chest pain,
characterized by reversible ischemia, and relieved by rest. Unstable angina will be defined as

pre-existing anginal chest pain that is increasing in severity and/or frequency. Myocardial

infarction will be defined as myocardial ischemia that progresses to myocardial cell death, with
subsequent liberation of intracellular cardiac enzymes, and varied clinical manifestations.

Chest Pain Quality

Pain is a personal, subjective experience, and can be expressed by one person to another
orally, but the experience can never be transferred directly. Webster's dictionary (Neufeldt &
Sparks, 1990) defines pain as the "physical or mental suffering caused by injury, disease, grief,
anxiety" (p. 422). This dictionary defines quality as "that which makes something what it is; the
characteristic element, basic nature, kind" (p. 481 ). Verbrugge and Steiner (1981) discuss
quality in terms of seriousness, finding that when the physician perceives a health problem as
serious, there is an increase in services and dispositions for follow-up care.

This researcher defines chest pain quality as the type of personally experienced
anginal-type pain that can be described by adjectives or adverbs for comparison.

CKMB

Creatine kinase (CK) is an enzyme specific to brain, myocardium, and skeletal muscle
cells. CK has three slightly different molecular forms, called isoenzymes. In this study we are
only interested in creatine kinase-MB (CKMB), found predominantly in the myocardium, the only
tissue containing sufficient CKMB to account for plasma increases (Underhill, Woods, Sivarajan
Frolicher, & Halpenny, 1989).

During an Ml, destruction of cell membranes occurs, releasing CK and CKMB into the
blood. The first detectable lab abnormalities are elevation of CK and CKMB levels.

Marin and Teichman (1992) state that "confirmation of the diagnosis of infarction depends
ultimately on the detection in blood of creatine kinase isoenzyme patterns indicative of
myocardial necrosis" (p. 354). Mair et al. (1991) found that CKMB mass was the best diagnostic
measurement for MI. Roberts (1984) reviewed Ml diagnostic practices, and found agreement
that CKMB is the most sensitive and specific diagnostic marker, and has become the
conventionally accepted hallmark of acute MI. Sensitivity and specificity of CKMB will be

discussed during the literature review.

This researcher defines CKMB as an isoenzyme of CK specific to myocardial cells, which
is released into the bloodstream during myocardial injury, and is diagnostic for Ml when found in
elevated levels.

Other Variables

This researcher anticipates that older age, socioeconomic status (reflected by health
insurance status), marital status and social support, and delay in care-seeking behavior may
affect test outcomes, and could potentially become nuisance variables in the analysis of verbal
descriptions of chest pain and CKICKMB scores. Patients may delay care-seeking for chest
pain, potentially increasing the severity of their disease outcome and their CKMB values for the
following reasons: fear of the high cost of ED services and inadequate insurance coverage; and
lack of social support encouraging care—seeking behavior (indirectly reflected in marital status).

Literature Review

The relationship between chest pain quality and gender and, secondarily, CKMB results
brings up several subtle issues that need consideration. This literature review begins with a
survey of CHD, and treatment modalities, as related to gender. Subsequent areas for
exploration include: care seeking behaviors of men and women, and their perceptions and
expression of illness; perceptions of care providers to client presentation (male or female);
clinical decision making processes; and the sensitivity and specificity of a CKMB result as a
diagnostic tool.

Sociodemographics of CHD

CHD is the leading cause of death in men and women in the United States. CHD affects
about 7 million Americans, causing over 500,000 deaths annually, and killing as many people as
all other diseases combined. CHD also costs taxpayers roughly $43 billion per year in direct and
indirect costs, and is one of the leading causes of disability (US. Department of Health and
Human Services, 1990). Men had a higher incidence rate within all age groups, and were more
likely to be first diagnosed with an acute form of CHD, whereas women were more often

diagnosed with chronic CHD (Center for Disease Control, 1992). Investigators agree that

women with Ml are typically older, and have a greater overall mortality rate (Greenland et al.,
1990; Tofler et al., 1987).

Lack of social support is reflected in a 3—fold increase in mortality rates from CHD, in
women (Eaker, 1989). Older women experience dramatically higher rates of singleness due to
loss of spouse thorough death or divorce, and having never married. Three-fifths of women over
age 65 are without spouses, whereas three-fourths of men in this age group are married (Jecker,
1991)

Clinical manifestations of MI can vary widely, but typically the patient presents with severe
anterior precordial chest pain of a pressing or squeezing nature. This pain is not relieved by rest
or nitroglycerine, and often radiates to arms, neck, or jaw. Physiological changes include
nausea and vomiting, perspiration, weakness, extreme anxiety, dyspnea, and a subjective sense
of impending doom (New York Heart Association, 1979). We must keep in mind that patients
with an Ml that do present for evaluation are the "survivors", as life threatening ventricular
arrhythmias and sudden death before reaching help is a common complication of infarction.

MI may occur, but not be associated with typical anginal symptoms. These are considered
"silent" infarctions, and may go unrecognized. "Silent" infarctions carry the same prognosis for
long-tenn survival as recognized infarctions (Hedges 8. Kobemick, 1988).

Eysmann and Douglas (1992) discuss patient risk in terms of their prognosis and poorer
outcome; severity of disease process and development of complications; and incidence of
mortality found with AMI. Governmental agencies speak of risk as personal attributes or
behaviors that increase or decrease the likelihood of developing a disease, morbidity and
mortality, or disability. Some risks are identified as modifiable, i.e., smoking, obesity, and
sedentary lifestyle (Center for Disease Control, 1992; Horton, 1992; US. Dept. of Health &
Human Services, 1990).

Plotnick and Fisher (1985) define high-risk for AMI as the presence of a true pathogenic
process like coronary thrombosis, coronary spasm, and/or progression of disease. Conversely,

Pozen et al (1984) defines low-risk as a decreased likelihood of having AMI. In a study of

low-risk patients in the ED, Lee et al. (1985) determined low-risk by the combination of
verbalizations of a "sharp" pain, chest wall pain that was positional, reproducible, pleuritic, and
no prior history of angina.

Lerner and Kannel (1986) reviewed data from the Framingham Heart Study, initiated in
1948 to identify coronary risk factors in a community setting. They found that overall women
experience half the amount of CHD as men, until old age. However, there was a 10-fold
increase in CHD when comparing younger women (35 to 54 years) to older women (>55 years).
Male risk for CHD increased only by a factor of 4.6 during the same age span.

The F ramingham data, often considered the "gold standard" for cardiac studies, has come
under criticism regarding the credibility of its epidemiological data, specifically the perception
that angina pectoris is a benign problem in women. Lerner et al. found that anginal chest pain is
a more common presentation in women than in men reporting chest pain. Eighty-six percent of
angina in women was uncomplicated by coexistent MI, whereas male angina is more likely to be
a sequelae of MI (66% uncomplicated). Men had a higher incidence of MI overall, and sudden
death from MI, while women were found to have a higher mortality than men during an Ml at
every age level.

The Framingham data have shown that the clinical manifestation of a similar
pathophysiological problem, coronary atherosclerosis, resulted in quite different outcomes in
men and women. Wenger (1990) voices concern that these results betray a bias in treatment.
She states:

"This flawed myth of better tolerance of angina fostered less attention to women

with this symptom, less concern with their preventive care and coronary risk

modification, and probably led to inappropriate decisions about objective testing

for risk stratification with resultant lack of identification of high-risk women..." (p. 557).

The US. Public Health Service's Task Force on Women's Health Issues (Kirschstein, 8.
Merritt, 1985) reports that treatment for women is based on research results largely established

from male study subjects. Application of results, (in the form of treatment protocols), overlooks

that women may react differently to treatments from men, and that some diseases manifest
themselves differently in women (Council on Ethical and Judicial Affairs, AMA, 1991 ).

Douglas (1986) believes that the differences in male and female responses render
published results liable to inaccuracy regarding applicability of findings to all individuals. She
sees gender-based research as a way to improve understanding of the pathophysiology that is
the underlying basis of much health care practice.

Gender Differences in Treatment for AMI

Is there a gender bias in the diagnosis and treatment of AMI? When we consider clinical
presentation for chest pain, men and women should receive the same diagnostic procedures,
effective therapeutic services, and satisfactory dispositions for follow-up care. Verbrugge and
Steiner (1981) believe that physicians rely on other factors that may prompt different procedures
and care: general knowledge about disease prevalence, risk factors, and physiology by sex.
Social and psychological effects may also influence care. Unless grounded in physiological
differences, this represents biased care.

Studies of the differences in referral for coronary procedures found that women undergo
fewer major diagnostic and therapeutic procedures than men (coronary angiography,
angioplasty, bypass graft surgery). Researchers have speculated that this may represent either
underuse in women or overuse in men (Ayanian & Epstein, 1991; Schmidt & Borsch, 1990).
Women were more likely to require emergent cardiac surgery, experience higher operative
mortality and perioperative infection, less relief of symptoms, and greater reocclusion rates with
these procedures (Hawthorne, 1994). These effects may be attributed to later onset of disease
in women by 10-20 years, poorer functional status and comorbidity on presentation, and smaller
vascular physiology (Herman, 1993). Bickell et al. (1992) conclude that women were less likely
to be referred for graft surgery among low-risk patients, but shared equal likelihood with men for
referral as disease severity increased.

Exercise stress testing is less accurate in prediction of CHD in women. Even when results

were abnormal, Tobin et al. (1987) found that men were 10 times more likely to be referred for

10

angiography. The difference in referrals was not explained by patient history, symptoms, or test
results. Physicians considered anginal symptoms in women with abnormal test results to be
more likely related to non-cardiac causes than they did in men, even though women were more
symptomatic than men. However, an interesting note is that there was no difference in the
prescription of antianginal medications.

Tissue plasminogen activator (TPA) is one of the newer drugs available for thrombolytic
therapy during an Ml. TPA, when given to eligible patients promptly on arrival to the ED, aids in
reperfusion of ischemic myocardium by dissolving existing clots occluding coronary vessels.
Cardiac muscle becomes reoxygenated, saving tissue and decreasing death from arrythmias.
The challenge of TPA, and the other thrombolytics, involves the timely identifiwtion of patients
experiencing MI, and determination of eligibility for therapy. TPA therapy is most beneficial
when initiated early, preferably within six hours of chest pain onset.

Two intriguing studies involving TPA therapy show characteristics related to gender. An
analysis of factors causing prehospital delay in seeking care and obtaining appropriate
thrombolytic therapy found that four variables were predictors of increased prehospital time.
These were: slow symptom progression; low income; female gender, and being elderly (Schmidt
& Borsch, 1990). Patients surveyed most frequently cited perception of their symptoms as "not
serious" as the reason for not seeking care sooner. Hawthome's (1994) qualitative study of
gender and coronary surgery found that women perceived cardiac illness as an "expected" life
event, compared to men, who perceived it as a crisis event. Women may be minimizing the
seriousness of symptoms. It is possible that this may reflect the generally later onset of CHD in
women. The older one gets, the more "expected" is any illness. In fact, this researcher
contends that each of the four variables is pertinent to a woman's prehospital delay.

Maynard, Althouse, Cerqueira, Olsufl<a, and Kennedy (1991) looked at the role of gender
in TPA eligibility, and its subsequent administration. They found that men successfully met
eligibility criteria more frequently than women, and that men and women differed in their

ineligibility criteria. As expected, the upper age limit of 75 years was exceeded more frequently

11

by women. When eligible, women received TPA less often (55%) than men (78%). The study
could not answer why treatment was not given to women when eligible, due to lack of information
on the patient record.

It appears that diagnostic and therapeutic treatments (and treatment efficacy) for AMI
show gender bias, which in some instances is unwarranted. Greenland et al. (1991) summarize
these studies when they recognize that women who develop Ml are at greater risk for mortality,
and deserve aggressive assessment and treatment, including thrombolysis.

Perception of Illness, and Care-Seeking Behaviors

Gender differences in utilization of health care have been hypothesized as due to
differences in the way symptoms are perceived, evaluated, and treated (Hibbard & Pope, 1986).
This researcher contends that one's perception of symptoms affects the description of symptoms
to others, and one's care-seeking behaviors. The literature showed that several factors
contribute to medical care utilization rates, but were inconsistent in conclusions regarding health
perception and medical care utilization. Avis, Smith and McKinlay (1989) studied a person's
beliefs about their susceptibility to a condition or disease. Research has shown that people tend
to underestimate their own risk of developing certain conditions, or diseases, termed "optimistic
bias", and that perceived susceptibility is essential in motivating alterations in behavior. Their
results showed that in the case of CHD, self-perceived health status, along with established
epidemiological risk factors, influences a person's perception of their risk for CHD.

Hibbard et al. (1986) postulate that women have a greater interest in health care in their
role as "family nurse", thus making health issues more important to them. These authors
conclude that a positive correlation exists between women's symptom perception, and interest in
health, with higher medical utilization rates, especially for mild illness, due to low stigmatization
for such actions.

Verbrugge and Steiner (1981) looked at patterns in medical utilization by gender, finding

that women received more total services, dispositions for follow-up care, and minor diagnostic

12

services. Men received more major diagnostic services (ECG, X-rays, office surgery),
counseling, and referrals to specialists.

Hawthorne (1994) identified themes in the perception of cardiac illness that are related to
gender. Men regarded cardiac illness and events as a frightening confrontation with mortality,
which stimulated a life reassessment, not unlike a "mid-life crisis". This makes sense when we
remember that CHD manifests itself in the middle years in men. Men experienced a change in
perspectives and priorities. Illness was seen as a tangible, and hence manageable phenomena.
Men tended to search for causative factors of illness, and may experience an element of
self-blame.

Women in Hawthome's study (1994) tended to minimize symptoms, and the possibility of
CHD. Women's cues to disease reflected an "otherness" orientation to relationships and family
needs coming first, before attending to health problems. These results are in contrast to
Hibbard's conclusions discussed previously.

Women reported having a personal physician with greater frequency, and longer waits to
get an appointment time. This may reflect a pattern of routine use of services for primary care,
and the establishment of a stable client-physician relationship. Some women are apt to use an
obstetrician or gynecologist for their primary care, developing a history of ongoing care (Grimes,
1988).

The literature does not agree on the issue of physician perception of client behavior,
raising questions about whether there is a difference in the way physicians perceive behaviors
and symptoms of male and female clients. Physician gender may have an effect on study
results, but because there is only one female ED physician at the study site, the available data
for comparison of physician gender, perceptions, and interactions with male and female patients
preclude addressing this issue in the current study.

Clinical Decision-Making

Clinical decision-making is an important part of the identification and treatment of patients

presenting with AMI. The goal of decision-making is appropriate risk stratification of patients to

13

take advantage of therapy to reduce myocardial necrosis, and therefore morbidity, while at the
same time screening low-risk patients to save resources and health care dollars (Alonzo, 1986).

Decision-making should be based on the interactions between biomedical knowledge,
rigorous problem analysis, weighing of probabilities, and acceptance of risk. These interactions
can be influenced by clinician-patient interactions and sociocultural setting (Eisenberg, 1979).
Desicion-making is difficult clinically, since it usually happens quickly, relies on large quantities
of data, is mostly unconscious, and differs in style for different clinicians and presenting
complaints.

Researchers have tried many strategies to improve the accuracy of emergency
department triage and diagnosis of AMI. These strategies range from the identification of
high-risk clinical indicators, and rapid access to laboratory studies (Marin & Teichman, 1992), to
computer-based predictability protocols (Goldman et al., 1988), and 12-lead or 18-lead ECGs.

Various studies describe the high sensitivity that ED physicians have for identifying and
admitting Ml patients. Tierney et al. (1986) states that "implicitly or explicitly, physicians deal
constantly in probabilities. ...one essential clinical skill is the ability to accurately estimate the
probability of disease in individual patients..." (p. 12). This has been achieved at the expense of
overdiagnosing and over-admitting low-risk patients (Lee et al., 1985; Pozen et al., 1980; ). In
the US, only 50% of the patients admitted to the coronary care unit (CCU) ultimately are
diagnosed with AMI (Pozen, D'Agostino, Selker, Sytkowske 8. Hood, 1984).

The opening interview between care provider and client has been regarded as a critical
time for negotiation between patient and physician concerning which verbal descriptions of
symptoms and illness experiences are to be considered as serious, and addressed during that
visit (Roter, Lipkin & Korsgaard, 1991). Diagnostic accuracy may be enhanced by careful
attention to patient presentation and problem description, which may highlight patterns
suggestive of high or low risk for AMI. Hedges and Kobemick (1988) reflect that clinicians must
base clinical impressions on patient risk factors, presence of clinical abnormalities,

circumstances that prompted care-seeking behavior, and on the degree of patient discomfort

14

(chest pain quality), and the patients' description of the discomfort. Hawthorne (1994)
promotes the nurse as an effective bridge between caregivers, and patients, as nurses have long

been accustomed to translating patient vernacular language into clinically meaningful language

that could be used in pattern recognition.

CKMB as a Diagnostic Tool

CKNIB accounts for 15 to 22% of total CK activity in diseased myocardium. Elevation of
CKMB occurs rarely in other disease processes, but has been related to muscular dystrophy,
hypothermia, alcohol overdose, Reye's syndrome, and poisoning. It may also be associated with
heavy exercise (Gibbler et al., 1990; Underhill et al., 1989). CKMB activity can be documented
in the serum 4-8 hours post-infarct, and follows a predictable rise and fall over a 3 day period,

usually peaking at 24 hours after onset of chest pain (Lee & Goldman, 1986; Underhill et al.

1989). This is illustrated in Figure 1.

 

 

_ CK
-_-. CK MB
_ AST
I \ — — LD
l t‘ - LDH-I
’ I
I I
I I
I I
I
, I
I I
' I --_
, t .7‘\. ‘
I ' ' \‘ ‘
I ' \, ‘
,' i/. \,
' IVII \\ \
I I. \. ".
I It.“ \‘\ ..s...
' : L, '1 A A L .1 ‘\‘ A ‘L-‘~‘u A
I 2 3 4 5 6 7 a 9 to II 12 3

DAYS AFTER ACUTE MI ONSET

Figure 1. Patterns and timing of elevation of creatine kinase (CK), creatine kinase-MB (CKMB), and other cardiac enzymes.
(Adapted by Underhill et al., ( 1989), from Ravel, R., (1984). Wtow medicine: Clinical application of laboratog

_<_Ia_ta (4th ed.).)

15

The literature evaluates two diagnostic issues: 1. what are the specificity and sensitivity
for CKMB; and 2. how frequently should CKMB be sampled?

Sensitivity of CKMB is defined as all patients with both MI and elevated CKMB levels,
divided by all patients with MI. Sensitivity shows the true positive rate in patients with Ml.
Specificity is defined as all patients both without MI and with a negative CKMB level, divided by
all patients without Ml. Specificity is the true negative rate in patients without infarctions (Lee &
Goldman, 1986).

Quantitative studies have demonstrated that elevation of serum CK and CKMB
concentrations are the most sensitive test available for the diagnosis of AMI; they have also
shown acceptable levels of specificity for documenting Ml (Fisher et al., 1983). CKMB results
show sensitivity of 95% to 98%, and specificity of 92% to 95%, when compared over a 24-hour
period of serial measurements. (Dufour, 1987; Gibler et al., 1990; Lee & Goldman, 1986).

It needs to be kept in mind though that CKMB activity does not increase in serum until 4-8
hours after chest pain onset. This decreases sensitivity in the early stages of AMI, e.g., typically
when the patient first presents to the ED, and has an initial CKMB level measured (Mair et al.,
1991). Experts agree that a single measurement of CKMB in the ED shows a low specificity for
exclusion of MI. Therefore serial CKMB measurements are recommended over the first 24 hour
period after chest pain onset to increase diagnostic accuracy (Irvin, Cobb & Roe, 1980; Lee et
al., 1987; Lee & Goldman, 1986; Marin & Teichman, 1992). Falsely negative results are rare,
but may occur from inadequate sampling frequency (i.e., only once in 24 hours), or from
sampling that is too late after infarction (when levels have returned to normal). The study site
generally follows these recommendations for most patients admitted with AMI by measuring
CKMB levels at 6 hour intervals in the first 24 hour period. The study site does not perform
outpatient CKMB studies in follow-up for patients discharged from the ED.

In conclusion, this researcher believes that existing research has several strengths, but
also shows several shortcomings. The literature review showed a general interest in exploring

issues surrounding gender and cardiac health. It is evident that research efforts are beginning to

16

plumb the depths of the knowledge gap as far as effects and efficacy of coronary diagnostic and
therapeutic measures for women go. Sociodemographic studies thoroughly discussed the
incidence of CHD in the general population. An over-reliance on the F ramingham Heart Study
data was noted.

Health perception issues were interestingly covered from both patient and physician
perspectives. However, physician perception data were based on ongoing primary care
situations, rather than on perceptions of patients in crisis situations, as we would expect with
AMI.

CKMB testing and efficacy were covered at great length in the literature, having been
subjected to research for the last two decades. Refinement in methods, and in accuracy of
results was evident.

Except for Hawthome's article (1994) summarizing qualitative nursing research on
post-operative cardiac patients, the general lack of literature addressing theory-based nursing
issues and roles in the care of potential cardiac patients was dismaying. Advanced practice
nurses do care for cardiac patients, and staff nurses in primary care or the ED are often involved
in triage of cardiac patients. It appears that nurses either lack recognition of these roles, or are
afraid to claim this expertise for themselves and acknowledge it as being within the realm of
their practice.

Nevertheless, given the state of the literature, it was obvious that the process of patient
history assessment, and clinical decision-making was downplayed in the diagnosis of AMI, when
compared to cardiographic and laboratory data. Three articles briefly mentioned patient
presentation and chest pain quality. One author conducted research using patient verbalization
to identify low-risk patients, but results were not gender specific, and an explanation of
methodology and verification of results was missing.

This researcher believes that assessment of patient presentation provides a contact with
the patient that is rich with information and clues relevant to the etiology of their chief complaint.

Current literature does not address the questions of possible gender-based differences in chest

17

pain description, and predictability of severity of disease based on patient verbalizations of chest
pain quality. The present research study addresses these issues with the intent to strengthen
assessment and history-taking as useful tools in the triage and clinical-decision-making process
for patients experiencing chest pain.

Methods

Research Design

This study was based on a non-experimental, retrospective correlational design. This
researcher did not have manipulative control over the independent variables, as the events
under study had already taken place. According to Polit and Hungler (1991), correlational
studies look at interrelationships between two or more variables, and the tendency for variation
in one variable to be related to variation in another.

The independent variable identified for the first question is the gender of the patient. The
outcome, or dependent, variable is the description of chest pain quality. This variable is
measured by recording descriptive adjectives used by patients of different gender, and
examining frequencies of their use.

The second research question is addressed by adding the descriptive adjectives as
independent variables in addition to gender, and treating diagnosis and hospital admission to a
critical care unit (yes, or no) as the outcome variables to be predicted.

The same two independent variables, gender of the patient and adjectives used to
describe chest pain, were employed in the analysis related to the third question. Here, the
outcome variable consisted of: 1. two creatine kinase (CK), and 2. two CKMB laboratory results
(initial, and peak or 24 hour values) for each subject in the sample admitted for AMI.

Additional independent variables were explored as control variables, including: 1. age of
patient; 2. marital status; 3. health insurance coverage (as a proxy for socioeconomic status i.e.,
Medicaid vs. others); 4. prior cardiac history; and 5. time from onset of symptoms to ED
presentation. Cardiac history and time to presentation are included because of their influences

on clinical decision-making and disposition.

18

Sample

Northern Michigan Hospital (NMH) in Petoskey, Michigan served as the study site. This
hospital is a regional referral center, serving 350,000 residents in 24 counties in the northern
lower peninsula, and eastern half of the upper peninsula of Michigan, and is available to
residents in Ontario, Canada. NMH is a 299-bed facility, with an Emergency Department offering
services on a 24 hour basis. Burns Clinic, which provides physician staffing for the ED, is the
largest physician-owned and operated professional corporation in the state, supporting over 125
physicians, including six emergency physicians, six cardiologists, and two cardiac surgeons.
NMH has 18 intensive care beds, and 30 cardiovascular beds. Two cardiac catheterization
laboratories, and open heart surgery facilities are available.

The target population of this study included all men and women presenting to the ED at
NMH with chest pain as their primary reason for seeking health care. A convenience sample
was used, taking all patients that fit eligibility criteria. Although the primary sample of interest is
women with suspected cardiac chest pain the study design requires the inclusion of eligible men
and their descriptions of chest pain to allow evaluation of gender differences.

Projected sample size for the study was based on a review of the ED census log for
January through March 1994. During these months, almost 200 patients were seen for chest
pain. Review of the ED census log, to identify eligible patients, took place during the months of
May, June, and August 1994. Northern Michigan experiences a large population increase during
the summer months, due to vacationers, and the return of retirees. This had the potential to
increase sample size.

Eligibility criteria for all questions included: men and women 30 years old and older,
presenting with chest pain that is not related to trauma, or radiographic abnormalities
(pneumonia, asthma, COPD). For the third research question, criteria also included admission
to a critical care area with a diagnosis of AMI, and availability of CKMB results from admission,

and at 24 hours.

19

Exclusion criteria included: 1. patients transferred from another hospital; 2. patients seen
by physicians other than the six regular physicians employed by the Burns Clinic to staff the ED;
3. patients who died during prehospital treatment (i.e., during resuscitation efforts by Emergency
Medical Services personnel).

Data Collection and Recording

Hospital records were obtained from the hospital's Medial Records Department. Data
gathering occurred through retrospective chart review either of the ED record for patients
presenting to the ED with chest pain who subsequently were discharged from the ED, or of the
ED record and inpatient hospital chart for admitted patients.

According to Hayward, McMahon, and Bernard (1993), "review of the medical record is an
integral part of evaluating the quality and appropriateness of inpatient care (p. 550). The Joint
Commission on Accreditation of Health Care Organizations has established guidelines regarding
basic information required in an ED record (Joint Commission of Accreditation of Hospitals,
1983). Guidelines pertinent to this study include proper documentation of patient identification,
time and method of arrival; pertinent history of illness; diagnostic and therapeutic orders; reports
of procedures, tests, and results; diagnostic impression; and final disposition. Data were
collected by one researcher, providing consistency and reliability in coding.

Ethics

No identifying characteristics were included in the data coding process, to protect patient
and provider confidentiality. Approval for this study was obtained from Michigan State University
UCRIHS, and from Northern Michigan Hospital's Institutional Review Board before conducting
the research (see Appendix A). Findings were forwarded to interested committees and staff at
NMH, as desired.

Operational Definition of Variables

Operationalization of the variables gender (male or female) and chest pain quality
(descriptive adjectives) has already been mentioned. CK and CKMB levels were measured on

serum blood samples obtained from each patient by venipuncture, on admission to the ED. The

20

NMH laboratory uses the DuPont Dimension AR to run samples through an immunoinhibition
enzymatic assay. Results are usually available within one hour. Follow-up of abnormal results
is done by Baxter Stratus pH fluorescence immunoassay. If results are still reported as
abnormal, due to interference, specimens are sent to the Burns Clinic laboratory for
electrophoresis chromatography.

Normal values for total CK range from 35—232 UIL in men, and 21 -21 5 UIL in women.
Normal CKMB results are 0-13 mgldl for both men and women. CKMB is also computed by the
laboratory as a proportion of the total CK result: (CKMB divided by total CK) x 100. Scores for
CK and CKMB levels were recorded for each patient. The NMH laboratory protocol for reporting
a CKMB result is available in Appendix B.

An admitting or ED diagnosis (for those sent home), and a discharge diagnosis for
admitted patients was coded for each patient. Patients with a discharge diagnosis of MI, or
unstable angina were considered "high-risk". Patients with a discharge diagnosis of angina
pectoris, and those sent home from the ED were considered "low-risk".

Data Analysis Procedures

Data analysis was performed with the help of a computerized statistical analysis program,
SPSSIPC+ (Norusis, 1992). Frequency data were generated for the demographic variables, and
for identification of the most commonly occurring adjectives used to describe chest pain quality.

The Pearson Chi-square, or the Fisher Exact tests, were used to evaluate whether
demographic variables showed any statistically significant interrelationships. The same analysis
strategy was also pursued with the first research question to determine relationships between
gender and chest pain description. Correlations were run on the most frequently used
descriptive adjectives.

Logistic regression analysis was chosen to analyze the multivariate problems addressed
in the second research question. This technique is appropriate when the outcome variable is
dichotomous, and the influence of multiple predictor variables on the likelihood of one or the

other outcome is to be investigated (lhlenfeld, 1988).

21

Multiple regression analysis was used for the third research question involving the issue of
predictability of elevated CKMB results. In these analyses the actual CKMB scores are the
outcomes to be predicted. Accepted levels of significance were .05 for all methods and
calculations.

3%

Demographics

Demographic data included patient gender, age, marital status, medical insurance status,
prior cardiac history, and time from symptom onset to presentation to the ED. One hundred
ninety-one people presenting to the ED with a primary complaint of chest pain were eligible for
inclusion in the study during the period of investigation. One hundred one (52.9%) patients were
men, and 90 (47.1%) were women. Sample age ranged from a minimum of 31 years, to a
maximum of 91 years, with a mean age of 61.8 years.

Marital status showed statistically significant correlations with gender and patient age.
Overall, 73.5% of the sample were married, 13.3% single, 2.8% divorced, and 10.5% were
widows or widowers. Comparing marital status by gender, 84.4% of men were married, whereas
only 61.2% of the women were married. Unmarried participants were more likely to be older, as
shown by their mean age of 62.44 years (std. dev. = 19.5 years), compared to those who were
married with a mean age of 60.91 years (std. dev. = 13.06 years.) No less than 75% of the men
in any given age category were married. As an example, women lagged 24.1%, and 50% behind
men for percentage of married participants among those 70-79 years old and those 80-89 years
old, respectively.

Health insurance status was divided into four categories: those who had private
insurance, Medicare, Medicaid, or no insurance. This information was used as a proxy for
socioeconomic status among people under the age of 65 since women are less likely to have
private health insurance, and more likely to rely on public assistance coverage, than men

(Horton, 1992). 73.3% of participants had private medical insurance, 51.8% had Medicare,

22

4.2% had Medicaid, and 2.1% had no insurance. Sixty-two people had both private insurance
and Medicare coverage.

Correlations between gender and insurance status were significant for private insurance,
in that 81.2% of men, versus 64.4% of women had this type of insurance. No significant
difference was demonstrated between gender and other types of insurance. Because of the low
Medicaid distribution, insurance status was not used as a variable in further analysis equations.

Prior cardiac history was also divided into four categories: patients with a history of
angina pectoris, unstable angina pectoris, myocardial infarction, or "other” cardiac history that
included such diagnoses as congestive heart failure, arrhythmias, cardiac surgery, valve
dysfunction, and pacemaker placement. 45.5% of those eligible had a prior history of angina
pectoris; 7.9% had unstable angina pectoris; 24.6% had a history of myocardial infarction; and
57.1% had other types of cardiac history. 33% of those presenting with chest pain had no prior
cardiac history at all, while at the other end five persons had cardiac histories covering all four
categories.

The time from symptom onset to patient presentation at the ED may be a factor in the
decision to use thrombolytic therapy in those patients experiencing a myocardial infarction.
Patients were evaluated to see if they presented in six hours or less from symptom onset, or in
greater than six hours. Six hours was chosen because older TPA protocols were less likely to
qualify patients as eligible for therapy if they waited longer than six hours to seek evaluation for
chest pain. In this study 47.1% presented within six hours, with 52.9% waiting longer than six
hours. Overall, there was no significant difference in time to presentation based on gender or
prior cardiac history, and this variable was not used in further analysis.

Admission statistics for diagnosis and age by gender are presented in Table 1.
Admission statistics were fairly consistent for men and women, by diagnosis. However, we see
that men were more likely to be admitted than sent home after the age of 50 years, while women
were less likely to be admitted than discharged from the ED until after the age of 60 years, that

is, ten years older than men.

23

Admitting diagnosis was compared to inpatient discharge diagnosis, by gender, and is
presented in Table 2. Obviously, 100 percent of the patients admitted with MI (8 patients)
correspondingly had a discharge diagnosis of MI. But, almost double that number of patients,
(15 patients) were determined to have had an MI by their discharge date. In men, 14.7%
admitted with unstable angina, and 50% (in this case, 1 patient) admitted with "other cardiac"
diagnosis had findings indicative of MI by discharge. 3.6% of women (again, 1 patient) admitted

for unstable angina subsequently had an MI.

24

Table 1. Admission Statistics for Diagnosis and Age by Gender

 

Admission Percent Percent
Statistics Admitted Discharge_d
Total 60.7 39.3
Gender Male 62 38
by Diagnosis
Angina Pectoris 45.9 54.1
Unst. Angina 100 0
MI 1 00 0
Other Cardiac 100 0
Unknown Cardiac 50 50
Non-cardiac 0 1 00
by Age
30-39 yr. 12.5 87.5
40-49 yr. 46.2 53.8
50-59 yr. 63.6 36.4
60-69 yr. 77.8 22.2
70-79 yr. 75 25
80-89 yr. 55.6 44.4
90-99 yr. 50 50
Female 58.9 41 .1
by Diagnosis
Angina Pectoris 58.3 41.7
Unst. Angina 100 0
MI 100 0
Other Cardiac 100 0
Unknown Cardiac NA NA
Non-cardiac 4.3 95.7
by Age
30-39 yr. 0 100
40-49 yr. 21.4 78.6
50-59 yr. 37.5 62.5
60—69 yr. 80.6 19.4
70-79 yr. 75 25
80-89 yr. 84.6 15.4
90-99 yr. 100 0

 

 

 

25

Table 2. Comparison of Admitting Diagnosis and Discharge Diagnosis by Gender

 

 

Discharge
Diagnosis
# of patients
(percent)

Ang. pectoris
Men
Women

Unstable Angina
Men
Women

Ml
Men
Women

Other cardiac
Men
Women

Unknown cardiac
Men
Women

Non-cardiac
Men
Women

Death
Men
Women

Column totals
Men
Women

Admitting Diagnosis # of patients (percent)

Ang.
pectoris

8(471)
11(52xn

0
1(48)

3(t76)
o

1(5tn
1(48)

5(2910
8(381)

17(281”
21(4oxu

Unst.
Angina

12(351n
5(T79)

12(351»
16(571)

5(147)
1636)

2(5tn
3(107)

3(81»
2(71)

0
1636)

34(567)
28(531”

MI

7(100)
1(100)

7(117)
1(19)

Other
cardiac

2(100)

2(3:»
2(31»

Row totals

20(331n
16(301”

12(201»
17(327)

13(217)
2(31»

5(81”
5(91»

_L—l
AA

tog:
VV

9(151»
10(191»

0
1(19)

60(100)
52(100)

 

26

 

Research Questions

Research was conducted to evaluate how men and women described the quality of their
chest pain during initial assessment in the ED. Thirty-nine descriptive adjectives were used by
eligible participants, and are presented in Table 3. The frequency statistics reflect that some
participants used more than one adjective when describing their pain. One hundred two people
used only one word; 52 used two different adjectives; 29 used three adjectives; 7 used four
adjectives; and one person used five different adjectives.

Adjectives were identified as "frequently used" if they occurred at least seven (3.7%) or
more times in the patient records. The ten most frequently used adjectives were "ache",
"burning", "dull", "heavy", "pain", "pressure", "sharp", "stabbing", "tight", and "weight". These ten
adjectives form the basis of investigation for the research questions.

Correlation analysis was done for these ten adjectives. Slightly positive correlations were
found between the words "heavy" and "weight", and between "sharp" and "stabbing", indicating
that those who said "heavy" or "sharp" were also somewhat more likely to say "weight", or
"stabbing", respectively. A small negative correlation was found between the word "sharp" and
the words "burn", "pressure", and ”tight". All correlations were smaller than 1 .27. Gender
analysis of all of these correlations showed no significant differences by groupings of words.

However, since for the majority of the patients only a single adjective was reported, these
correlations are based on small varying numbers of cases. Nonetheless, they indicate that,
among patients that used multiple descriptors of their experience, no identifiable clusters of
descriptors emerged. As a result, the following gender comparisons will focus on the 10

individual adjectives most frequently recorded.

27

Table 3. Descriptive Adjectives

 

 

Adjective
Ache

Ball in chest
Breathless
Burning
Clenched fist
Cold
Constant
Constriction
Cramp
Crushing
Deep
Discomfort
Dull
Flutter
Fuflness
Gaslike
Grabbing
Heavy

Hot flash
Hurt
Jabbing
Knifelike
Mild
Numbness
Pain
Pressure
Punched
Pushing
Sharp
Shooting
Sore
Squeezing
Stabbing
Tender
Throbbing
Tight
Twinge
Viselike
Weight

39
1
1

-¥ N
O

AwwémbNthN-‘d

N
O)

m-s—swww-A

58

49

ANNOAO)

3

I\)

\j—L—l

quency Percent

20.4
0.5
0.5

10.5
0.5
0.5

1
1.6
2.1

1

1
2.1
7.9
0.5
1.6
1.6
2.1

13.6
0.5
1.6
1.6
1.6
0.5
0.5
4.2

30.4
0.5
0.5

25.7
1.6
2.1
3.1
3.7

1
0.5

16.8
0.5
0.5
3.7

 

28

 

The Pearson Chi-square or F isher‘s Exact test was used to address the first research
question. The comparison of how men and women describe their primary symptom of chest pain
found significant differences in the use of three adjectives: "heavy", "pain", and "tight"; by gender
(see Table 4). Women were more likely to use the adjective "heavy" while men were more likely
to use the adjectives "pain", or "tight". No other significant gender differences were found among

the remaining adjectives.

Table 4. Adjective Use by Gender

 

 

Adjective Percent Percent Significance
3% Female _I.e_ve_l
Ache 18.8 22.2 0.5594
Burning 10.9 10 0.8408
Dull 7.9 7.8 0.9704
Heavy 6.9 21.1 0.0043"
Pain 5.9 2.2 0.0002'
Pressure 32.7 27.8 0.4627
Sharp 21.8 30 0.1942
Stabbing 3 4.4 0.5883
Tight 22.8 10 0.0183‘
Weight 3 4.4 0. 5883

 

 

The second research question asked whether descriptions of chest pain quality predicted
hospital admission to a critical care area, and if so, were the same adjectives predictive for men
and for women? Logistic regression was used to determine association between adjectives and
admission. Because other factors may influence the clinical decision-making process, the
variables for initial CKMB results, and prior cardiac histories were included in the analysis.

Two adjectives were found to be predictive of the likelihood of admission depending on
the gender of the patient (see Table 5). Men using the term "pressure" (signif. = .0248) were

more likely to be admitted to a critical care area than discharged home, as were women using

29

the term "heavy" (signif. = .0398). Other adjectives were not found to be predictive of hospital
admission. It was found that women with "other" cardiac history (arrhythmias, surgery, etc.),
were also more likely to be admitted than sent home. This variable did not predict hospital

admission among men.

30

Table 5. Prediction of Hospital Admission to a Critical Care Area

 

Variable

Ache
Burn
Dull
Heavy
Pain
Pressure
Sharp
Stabbing
Tight
Weight
Gender
CKMB1

Cardiac History
(A. Pectoris)

Unst. Angina
Ml
Other Cardiac

Number of
Eligible Cases

 

Si nificance Level

M
0.7531
0.7497
0.5087
*0.0156
0.2412
"0.0059
0.5047
0.5497
0.161
0.4757
0.9356
0.246

'0.0208
0.724

0.3331
*001

170

 

Male Fema_le
0.571 1 0.6592
0.9917 0.6593
0.7406 0.924
0.2057 ‘00398
0.869 0.9475
*0.0248 0.0737
0.4092 0.4877
0.3394 0.5835
0.9209 0.8556
0.1744 0.8934
constant constant
0.2858 0.0706
0.2789 0.0545
0.7897 0.8895
0.1 129 0.8407
0.1857 *0.0065
90 80

 

31

 

The third research question used multiple regression to analyze whether descriptive
adjectives as used by women and men were diagnostically accurate predictors of elevated
CKMB results, and therefore indicative of MI. A few extremely high CKMB outliers were
excluded from the analysis to decrease the skewness of the dependent variable, in an attempt to
normalize the curve of distribution values. Frequency studies showed that 75% of the sample
had initial CK values below 108 UIL, and peak CK values equal to or below 99 UIL. Initial and
peak CKMB values averaged below 4 mgldl for 75% of the sample.

There were no statistically significant regression coefficients for the descriptive
adjectives used by the group as a whole, and their initial CK and CKMB values. Significant
coefficients were found for gender as predictor of initial CK and CKMB values, with women
averaging 30.32 UIL below the men's total CK values, and 2.017 mgldl below male CKMB
values. There were no significant regression coefficients for the descriptive adjectives, or

gender, to predict peak CK and CKMB results. These results are presented in Table 6a.

32

Table 6a. Overall Prediction of Elevated CKICKMB Results

 

Variable Initial CK Initial CKMB Peak CK Peak CKMB
Signif. Signif. Signif. Signif.
8(2) _L_$v_e| E w E w E LfleJ
Gender -30.32 *0.0056 -2.02 10.0183 -80.78 0.2289 -13.27 0.0696
Ache 2.19 0.8691 -0.37 0.72 23.15 0.7863 -5.05 0.5879
Burn 0.52 0.9788 0.08 0.9568 -4.92 0.9572 13.1 0.2782
Dull -18.43 0.3289 0.44 0.7704 -53.54 0.5816 —7.63 0.5167
Heavy -15.2 0.315 0.55 0.6517 26.68 0.7512 3.81 0.6694
Pain -1.63 0.9512 -1.82 0.3845 21.37 0.8771 -5.5 0.7461
Pressure 7.38 0.5369 0.57 0.5449 43.23 0.5567 7.35 0.3493
Sharp 13.1 0.3616 064 0.5671 2.75 0.9765 -2.27 0.8199
Stabbing -35.36 0.1823 1.05 0.629 -112.3 0.665 -0.41 0.9911
Tight 5.75 0.6846 -1.8 0.1189 129.71 0.1391 -0.33 0.9692
Weight 43.62 0.1004 0.44 0.3645 -32.38 0.7843 -13.01 0.3902
% Variance
Accounted
For 11.2 5.4 10.4 8.6

 

 

 

(") B = Regression Coefficient

33

An interaction effect was found when regression analysis was done for peak CKMB
results and adjective use specifically by female or male gender (see Table 6b). Women who
used the word "heavy" had CKMB results 11.74 mgldl higher than women not using the term
(signif. = .0482). No such effect was found in the male subsample. Again, the contribution to

explained variation in the dependent variable by this variable remains very small.

Table 6b. Gender-Based Predicted Elevation of Peak CKMB Results

 

 

Variable _M_al_e Female
Signif. Signif.
8(2) L184 8 L_ex_e_|
Gender -24.08 0.1323 constant
Ache -19.82 0.3405 5.84 0.3904
Burn -0.845 0.7489 -8.49 0.3793
Dull -23.92 0.355 -1.99 0.8135
Heavy -3.38 0.8632 1 1.74 ‘0.0482
Pain -37.89 0.3101 -6.66 0.6119
Pressure -13.31 0.4377 -7.26 0.219
Sharp -25.1 0.2516 -6.14 0.4016
Stabbing 32 0.6921 4.41 0.8195
Tight -18.39 0.328 -7.41 0.2855
Weight -13.93 0.6754 -5.01 0.6087
% Variance
Accounted For 6.6 18.1

 

(") B = Regression Coefficient

 

Other Findings

Out of curiosity, this researcher compared the discharge diagnosis of MI with descriptive
adjective use. Fifteen patients were discharged with MI. Twenty percent initially described their
pain as an "ache", 6.7% as "burning", 6.7% as "dull", 20% as "heavy", 13.3% as "pain", 40% as
"pressure", 13.3% as "sharp", and 20% as "tight”. These percentages add up to more than
100% because some patients used more than one adjective. None of the MI patients used the
words "stabbing", or "weight".

Interpretation of Findings

There is some evidence that women and men use different terms in their subjective
description of chest pain on presentation to the ED. Women were more likely than men to use
the term "heavy"; men were more likely than women to describe their pain as "tight", or "pain".
Without interviewing each subject individually, it is difficult to conclude whether this difference is
due to a difference in perception of the pain, or to an actual difference in manifestation of the
disease and its symptoms in women and men, questions that were raised in the literature.

Chest pain quality description was marginally predictive of hospital admission to a critical
area, with women using different specific adjectives from men. Women using the adjective
"heavy" were more likely to be admitted than women who did not use this term, or than men in
general. Men using the adjective "pressure" were more likely to be admitted than men who did
not. Different adjectives were predictive for women and for men, with predictive strength higher
for women. It is interesting to note that the term "heavy", more likely to be used by women, is
also predictive of their hospital admission.

Correlations found between gender and initial CK and CKMB values are reflective of the
normal values used by the NMH laboratory for these tests, with women having a lower normal
CK limit. Specific adjective use was not found to be a diagnostic predictor of elevated initial
CKMB results, and therefore is not useful as a tool to predict AMI and severity of disease.

The adjective "heavy" was found to be predictive for abnormally elevated peak CKMB

results in women, but not in men. This term has consistently shown significance when used by

35

women through all three research questions. We can conclude that women who describe their
chest pain as "heavy" are more likely to require hospital admission, and to have abnormally
elevated peak CKMB results, indicative of increasing severity of disease process. While ED
nurses may not see elevated initial CKMB results in these women, they can anticipate the level
and type of nursing care necessary to provide initial treatment, and can initiate appropriate
patient care protocols. This is consistent with Benner‘s hypothesis that experience builds
paradigms for expertise in nursing practice. Expert nurses who have taken care of many patients
with chest pain build on past knowledge to make graded qualitative distinctions in severity of
chest pain, and use this knowledge in clinical decision making.

Admission statistics confirmed results reported in the literature (Herman, 1993; Verbrugge
& Steiner, 1981) that found that the onset of cardiac disease begins approximately ten years
later in women than in men. However, perhaps these statistics are a reflection of bias in risk
appraisal towards placing women in a lower risk category based on their gender (Lerner &
Kannel, 1986; Wenger, 1990).

Several factors discussed in the literature could influence findings related to the increased
numbers of Mls that were diagnosed by inpatient discharge, compared to those diagnosed in the
ED. Since CKMB rises to detectable levels 4-8 hours after infarct, patients presenting promptly
following onset of chest pain will not have elevated CKMB levels initially, but will peak after
admission, if their Ml evolves. Additionally, those admitted with prodromal unstable angina may
also progress to Ml, or a patient may have a new onset of angina and MI after admission.
Finally, we have to consider the question of whether the diagnosis was missed by the physician
on admission. The literature shows that single measurement of CKMB in the ED has low
sensitivity for exclusion of MI, and recommends serial measurement for accurate diagnosis. The
study site does not perform outpatient serial follow-up of CKMB for those patients discharged
from the ED, and may miss a borderline case that is actually an evolving MI. In fact, at least two
patients sent home from the ED returned within 24-48 hours with a continuing complaint of chest

pain, and subsequently were admitted with an Ml. Protocol changes reflective of standards as

36

defined in the literature allowing for follow-up outpatient CKMB studies within 12 to 24 hours
after a patient is seen and discharged from the ED could correct this problem.

Patients who were discharged after admission with a diagnosis of MI were inconsistent in
their descriptions of the same phenomena (Ml), with the exception of the use of the term
"pressure". Similar words were used to describe MI, angina pectoris, unstable angina, and other
cardiac problems.

Coming to the ED for any type of problem is often perceived by many as a crisis, let alone
coming in for a potentially life-threatening symptom like chest pain. Perceived crises initiate
stress responses that may initially cloud a person's judgment or perception. This researcher,
having worked as a nurse for six years in the ED at the study site, and having cared for many
patients with chest pain, has seen first—hand the confusion that some patients, especially the
elderly, experience when asked by care providers to qualitatively describe their chest pain. Care
providers attempting to assess chest pain quality may "offer" adjectives to patients who are
unable to articulate descriptions themselves, such as "does it feel like a pressure, or sharp?".
Suggesting adjectives to patients may alter study results as care providers try to grasp the
patient's perception of chest pain quality, and patients grasp for answers to questions about
symptoms that are hard to describe.

2m

Implications for Advanced Nursing Practice and Primary Care

Nurses working in advanced practice in such varied primary care settings as family
practice, internal medicine, and emergency nursing will frequently care for patients with chest
pain, chronic cardiac disease, or those at risk for developing cardiac disease. Nurses are often
the first health care provider that the patient interacts with. Nurses must have the clinical
knowledge to recognize high risk patients presenting with chest pain, to accurately assess the
situation in a holistic manner, and to manage patient care through clinical decision making and
the use of paradigms that initiates appropriate patient care. This insures the best possible

outcome for the patient, that is, preservation or salvage of cardiac muscle tissue.

37

This research study demonstrates that nurses cannot rely on a patient’s description of the
quality of their chest pain as an assessment tool for severity of disease. Patients experiencing
chest pain use several different terms to describe the same disease process, and use similar
terms to describe differing types of disease like angina and MI. While women and men may
describe their chest pain differently, and this description may lead to different outcomes
(discharge or admission), descriptions cannot be used as a "cookbook" method for predicting
which patients are at risk for elevated CKMB values, and for MI.

The "formal" role of nursing versus recognition that actual nursing practice may extend
beyond the usual boundaries of nursing practice in rapidly changing situations such as dealing
with patients with chest pain reflect clinical expertise gained over time. Nurses in advanced
practice meld the scope of their nursing practice and their focus on health and holistic patient
care with the medical model of disease recognition and treatment. APNs recognize disease, as
well as symptoms, and initiate nursing care that helps manage symptoms, and minimizes the
effects of disease.

APNs in primary care settings are skilled in assessment of history of the present illness
(HPI), and easily apply this technique to patients with chest pain. Their assessment includes:
the usual state of health, chronology of symptoms, location, quality, quantity, setting, aggravating
and relieving factors, associated manifestations, disabilities, and pertinent negatives. It is
obvious that patient description of chest pain quality is only one piece in the overall assessment
of chest pain. The APN can then collaborate or make the decision to refer the patient to the
nearest ED, to provide supportive care until transport can be arranged to a facility equipped to
care for cardiac problems, or to provide definitive care in the primary care setting if the pain does
not seem to be of a cardiac nature. A system for telephone triage and collaboration with a
physician is a must for APNs who may be practicing solo in a rural or off-site setting.

Physical examination usually is an ongoing process during history gathering for the patient
with suspected cardiac etiology, and often is begun by the nurse. Objective data gathered

during exam includes vital signs, skin signs (color, temperature, diaphoresis, edema), oxygen

38

saturation and pulmonary assessment, cardiac rhythm and sounds, ECG, weakness, and
reproducibility of pain. All of this is within the scope of nursing practice, and aids in clinical
decision making.

Benner had several ideas that are reflective of the results achieved in this study. She
believed that expert nurses learn to recognize subtle physiological changes, to make qualitative
distinctions about changes observed, and to communicate these changes in a collaborative
manner to the physician. Many times this is done before objective data are available (CKMB
values). Clinical knowledge is refined over time, and nurses use past experiences (caring for
other patients with chest pain) as paradigms to guide and manage rapidly changing patient care
situations, while incorporating patient care standards into the scope of their nursing practice.

Assessment of patient verbalizations of chest pain quality is a subtle skill involving clinical
judgment, and is often overlooked as nurses strive to learn the latest technological procedures.
Benner stresses that nurse educators must recognize the connection between assessment and
clinical judgment, and focus curriculum on this important aspect of skill acquisition. APNs need
to work to legitimize the history assessment aspect of their practice, to seek the opportunity to
function in this role, and to extend practical and theoretical knowledge by sharing these skills
with novice and beginning nurses through mentoring.

Implications for Current Literature and Future Research

The results of this study highlight five points in the current literature. The first three points
deal with the process of clinical decision making, and the last two with gender-based research.
First, results of this study give some support to the idea that diagnostic accuracy may be
enhanced by careful attention to patient presentation, and problem description, which may
highlight patterns suggestive of high or low risk for AMI. Health care providers cannot rely solely
on patient description of symptoms, due to the possibility of differing perceptions or
manifestations of the same disease process.

Second, this research supports the idea that clinical decision-making is an important part

of the identification and treatment of patients presenting with AMI, and is well within the realm of

39

the APN in primary care. Patient verbalization may cue the APN to the possibility of hospital
admission, specific to gender.

Third, Tierney et al. (1986) believed that one essential clinical skill is the ability to
accurately estimate the probability of disease in individual patients. Results of this study show
those patient descriptions of chest pain quality in this instance do not sufficiently strengthen the
probability for elevated initial CKMB values, and thus is not an indicator for severity of cardiac
disease (Ml).

This study showed that there were some differences in how women and men described
chest pain symptoms. Additionally, several gender-based differences were found and described
earlier in the sample demographics, and in related findings regarding admissions and diagnosis.
This researcher agrees with Douglas (1986), in that gender-based research can be a way to
improve understanding of pathophysiology that underlies the basis of much health care practice.
Differences in male and female responses may lead to inaccuracy if results are applied
indiscriminately to all individuals. Treatment for women based on research results largely
established from male study subjects overlooks that women may have differing perceptions and
presentations than men with similar complaints.

This study could serve as a springboard for future research in the areas of patient
perception of chest pain, and gender responses to treatment. This study was done in an ED that
regularly sees large numbers of patients with chest pain. Patients with non-cardiac etiologies for
their complaints were excluded from this study, but a broader study could be done to determine
whether chest pain quality descriptions can distinguish non-cardiac from cardiac etiologies.

Follow-up research could be done for patients admitted with chest pain. Discussions with
cardiovascular nurses have raised similar questions about how inpatients perceive chest pain
symptoms, and of differing responses by physicians to patient complaints of chest pain based on
gender

Cardiac rehabilitation nurses working with cardiac patients at the study site believe that

patients change how they describe their chest pain symptoms as they progress through

40

rehabilitation. This is probably due to the influence of the nurses themselves, as they work with
patients to help them accurately describe symptoms. Research could investigate how patient
descriptions change over time with exposure to cardiac rehabilitation education, and assess
whether patients become more homogenous in their descriptions, or more accurate in describing
differing cardiac manifestations like angina pectoris.

Research with patients undergoing cardiac catheterization opens the possibility of
verifying severity of disease with patient descriptions. Also, at the study site, Ml patients
receiving TPA therapy undergo cardiac catheterization after receiving drug therapy. Their
verbalizations could be compared to patients presenting with angina pectoris, or unstable angina
to determine if disease etiology really correlates with chest pain description, and if gender
differences have a physiological basis.

Finally, primary care providers could be studied to determine how or if recognition,
assessment and clinical decision making skills change when caring for female cardiac patients
before and after being provided with education from study results.

Limitations of Research

This research was subject to limitations that may affect applicability of results to a
generalized population of cardiac patients. First, a non-random, convenience sample was used.
This researcher recognizes that ethnicity, cultural background, age, pain tolerance level, and
education are some factors that may influence how pain is experienced, and expressed.
Comparing subjects with similar backgrounds that could take these factors into account may alter
test results. For example, someone with more education will have a larger vocabulary to choose
from when describing pain than someone with fewer years of schooling.

While it was hoped to have double the number of subjects in the study, sample size was
ultimately dictated by time constraints, and hospital Medical Records policies that made access
to patient records difficult at times, and decreased the number of records that could be analyzed
at any one time. Reliability of data was ensured by having one researcher encode data from

patient charts. However, this may incorporate a systematic bias into study results, limiting

41

validity. This could be bypassed by standardizing chart review processes, and training multiple
reviewers in objective review methods.

While male patients were more likely to use the adjective "pain", it is felt that this term may
really be a catch-all term that was used when physicians or nurses did not document patient
descriptions in a more accurate manner. In fact, when comparing physicians‘ and nurses'
charting on the same patient, it becomes apparent that patients may be describing their chest
pain in different ways to different care providers. This researcher has experienced this situation.
Since research results showed no significant correlations between groupings of adjectives, one
has to wonder if patient perceptions change; if patients become confused about quality
questions; if care providers are documenting accurately and consistently; or if patients are really
feeling several types of pain at once. This becomes an important methodological issue in that
one would need to standardize the procedure for recording patient verbalizations of chest pain.

The documentation issue could be resolved in several ways. Patients present to the ED
24 hours a day, and are cared for by a rotating staff of physicians and nurses. Initial interviews
between patient and care provider could be audio recorded to verify documentation with
verbalizations. Patients could be interviewed after admission, or by telephone if discharged, by
trained objective interviewers, providing more consistency in documentation. This would require
seeking permission individually from each subject to be part of the study group, which was not
necessary in the present study.

Patient responses could be standardized without care providers "offering" biased choices
of pain descriptors by printing up a listing of adjectives used by subjects in the present study.
This could be shown to patients during interviews about the description of their chest pain
quality, allowing them to use the list as a reference tool.

The issue of physician gender and perception of patient care-seeking behaviors is an
important one that could not be addressed in the present study. Physician gender affected the
length, content, and structure of patient visits in a study by Roter, Lipkin, and Korsgaard (1991).

Three similar studies (Bernstein and Kane, 1981; Greer, Dickerson, Schneiderrnan, Atkins, and

42

Bass ,1986; Verbrugge and Steiner, 1981) investigated client gender and physician treatment for
ambulatory care patients in primary care settings. Medical care was similar for both sexes in the
studies, but Verbrugge et al. found that men scored higher on both extent (depth of medical
history and examination, and number of diagnostic tests ordered); and content (proportion of
procedures deemed diagnostic for the complaint, that were performed). Both gender and client
verbal expressions of the complaint were rated as important factors in physicians’ response.

This could be addressed at a facility employing more female ED physicians than the study site.
Finally, a limited number of subjects had highly abnormal CK and CKMB results that
skewed mean and mode calculations, and normality of curve shape in the regression analysis for

the third research question. Results came more into line with expected findings after these
cases were factored out of the analysis.

Summary

There appears to be a difference in how women and men describe their primary symptoms
of chest pain on presentation to the ED. This difference may cue the care provider to the
possibility of hospital admission, but not to the probability of elevated initial CKMB results as an
indicator of severity of disease. The APN gains expertise in recognition of high-risk cardiac
patients by the use of paradigms. Knowledge about how cardiac patients describe their chest
pain, and the possible implications of these descriptions strengthen the assessment and
diagnostic component of the APN role, and is a skill that can be acquired through education, and

collaboration on clinical judgment with other care providers, and shared through mentoring.

43

APPENDICES

APPENDIX A

Study Approval Letters

44

)7

1‘

‘f‘wl? ‘

OFFICE OF
RESEARCH
AND
GRADUATE
STUDIES

University Committee on
Research Involving
Human Subjects
(UCRIIIS)

Michigan Slale Unwersdy
225 Adnnnsfiuhonfiuflduw
East Lansmg Michigan

48824-1046
517/355-2180
FAX 517l’336-ll7l

 

MICHIGAN STATE

 

Ll Al I V' E

It S I I" Y

May 27, 1994

TO: Barbara C. Jaquith RN

8274

W. Stutsmanville Rd

Harbor Springs, Mi 49740

RE: IRBI:

94-254

TITLE: CHEST PAIN QUALITY DESCRIPTION As A PREDICTOR OF
ACUTE MYOCARDIAL ISCHEMIA: Is THERE A GENDER
DIFFERENCE?

REVISION REQUESTED: N/A

CATEGORY: 1-E

APPROVAL DATE: 05/27/94

The University Committee on Research Involving Human Subjects'(UCRIHS)
review of this project is complete. I am pleased to advrse that the
rights and welfare Of the human subjects appear to be adequately
protected and methods to Obtain informed consent are appropriate.

herefore,

the UCRIHS approved this project including any revision

listed above.

RENEWAL :

REVISIONS:

PROBLEMS/
CHANGES:

UCRIHS approval is valid for one calendar year, beginning with
the approval date shown above. Investigators planning to
continue a project beyond one year must use the green renewal
form (enclosed with t e original agproval letter or when a
project is renewed) to seek u date certification. There is a
maximum of four such expedite renewals possible. Investigators
wishing to continue a project beyond that time need to submit it
again or complete reVIew.

UCRIHS must review any changes in procedures involving human
subjects, rior to initiation of t e Change. If this is done at
the time o renewal, please use the green renewal form. To
revise an approved protocol at any other time during the year,
send your written request to the UCRIHS Chair, requesting revised
approval and referencing the project's IRB # and title. Include
in your request a description of the change and any revised
instruments, consent forms or advertisements that are applicable.

Should either of the followin arise during the course Of the
work, investigators must noti UCRIHS promptly: (1) problems
(unexpected Side effects, comp aints, etc.) involVing uman
subjects or (2) Changes in the research environment or new
information indicating greater risk to the human sub'ects than
existed when the protocol was previously reviewed an approved.

If we can be of any future help, please do not hesitate to contact us
0.

at (517)355-2180 or FAX (517)3

  

171.

avid B. Wright, Ph.D.

UCRIHS Chair

DEw:pjm

cc: Manfred Stommel

45

 

Northern Michigan Commlllo'd m lntemur caring
"“3 Hospital

g
.rII

...i ill

III

August 2. I994

Ms. Barbara Jaquith, RNC, BS
Department of Emergency Medicine
Northern Michigan Hospital
Petoskey, M I 49770

RE: "CHEST PAIN QUALITY DESCRIPTION AS A PREDICTOR OF ACUTE MYOCARDIAL
ISCHEMIA: IS THERE A GENDER DIFFERENCE”

Dear Ms. Jaquith,

At their July 20, I994 Meeting, the Northern Michigan Hospital Institutional Review Board voted unanimously to
approve the above referenced new study.

Approval of the above named protocol is for one year and is contingent upon the investigator fulfilling
responsibilities as outlined in the Code of Federal Regulations, Title 2i, to include:

I. Prompt reporting to the I.R.B. of changes in the research activity.

2. Prompt reporting to the [RB of unanticipated problems involving risks to human subjects or
others.
3. Obtaining I.R.B. approval for protocol changes before initiating those protocol changes, except

when necessary to eliminate apparent immediate hazards to human subjects.

/4-

W. Schuil, M.D., PhD.
.hairman, Institutional Review Board

  
 
   

Sincerely,

DWS/sh

Institutional Review Board Members:
Duane W. Schuil, M.D., PhD.
Mark .I. Bielaczyc. M.D.
Gary H. Shaw, MD.
David L. Behling, Chaplain
Connie Fisher
Nick Michels, MS, DNSc, RN
Bob Wilcox, RPh

III) (Lurirlaillli‘ Ali-rule
I’rluxlws. “II‘IIIEJII 19770-3207

All. 21.“,Illlltl

46

APPENDIX B

CHEMISTRY DEPARTMENT TECHNICAL MANUAL-CKMB
Clinical Laboratory, Northern Michigan Hospital, Petoskey, Michigan 49770
TITLE: CKMB TESTING PROTOCOL
PROCEDURE/POLICY NO.: CHE/CKMBMAN

PREPARED: J. Stout DATE PREPARED: 10I25/93 DATE ADOPTED: 10/25/93

INTRODUCTION:
The CKMB procedure used on the DuPont Dimension AR is noted by the company to
be the best utilized as a screening technique for the presence of the creatine kinase
MB isoenzyme activity in serum. Interference is a common problem using this
technology for CKMB testing and this policy is written to define what criteria to use in
reporting and in follow-up of abnormal results.

SPECIMEN REQUIREMENTS:
1. Serum specimens are required. Blood collection tubes containing EDTA, lithium
heparin, sodium fluoride or potassium oxalate should not be used. Corvac and SST
collection tubes do not affect the CKMB method.
2. Hemolysis will cause a false elevation of CKMB results and hemolyzed
specimens should not be used.

CAUTION: BIOHAZARD PRECAUTIONS MUST BE FOLLOWED WHEN DOING CKMB
TESTING.

PROCEDURE:
A total CK level should be run with each CKMB level. If the total CK is normal and the
CKMB result is abnormally high the results are suspicious of interference. When this
occurs please use the following testing algorithm.

TESTING ALGORITHM:
If the CKMB result is:
1. 0-13 mgldl ....... report DuPont Dimension AR result.
2. = or > 14 mgldl ....... check total CK on DuPont AR.

If the total CK result is:
1. >232 UIL ........ report AR result for CKMB.
2. = to or < 232 UIL ........ repeat CKMB on Stratus.

If the Stratus CKMB result is:
1. < or = 4.7 nglml ........ report "See Text" and enter Stratus results and normals,

i.e..: "CKMB by alternate method = _"

"Normals < 4.7 nglml.

2. > 4.7 nglml ........ Electrophoresis. Result with "SEE" and coded comment "Referred
to pathologist, additional report to follow if indicated". Also add in procedure free text
that the total CK is normal but the CKMB shows interference by routine and alternate
methods.

47

REFERENCES

REFERENCES

Alonzo, A A (1986). The impact of the family and lay others on care-seeking during life

threatening episodes of suspected coronary artery disease. Soial Sciencg Medicine ,2_2,

 

1297-131 1 .

Avis, N. E., Smith, K. W., & McKinlay, J. B. (1989). Accuracy of perception of heart attack risk:
What influences perceptions and can they be changed? American Journal of Public Health,
19, 1608-1611.

Ayanian, J. S., & Epstein, A. M. (1991). Differences in the use of procedures between women
and men hospitalized for coronary heart disease. New England Journal of Medicine, g,
221 -225.

Bernstein, B., & Kane, R. (1981). Physicians' attitudes toward female patients. Medical Care,

 

19, 600-608.

Benner, P. (1984). From novice to expert. Menlo Park, CA: Addison-Wesley.

Bickell, N. A, Peiper, K S., Lee, K. L., Mard, D. B., Glower, D. D., Pryor, D. B., 8. Califf, R. M.
(1992). Referral patterns for coronary artery disease treatment: Gender bias or good clinical
judgment: Annjals of lntemglJVleriicine, 116, 791 -797.

Bullock, B. L. & Rosendahl, P. P. (1984). Pathoghysiolggy: Adaptations and alterations jg
m. Boston: Little, Brown.

Center for Disease Control (1992). Coronary heart disease incidence by sex-United States,
1971-1987. Morbidity and Mortfltv Weekly Report, Q (29), 526-529.

Council on Ethical and Judicial Affairs, American Medical Association (1991). Gender disparities
in clinical decision making. LAM, 26_6, 559-562.

Douglas, P. S. (1986). Gender, cardiology, and Optimal medical care. Circulation, 711, 917-919.

48

DuFour, D. R., LaGrande, A., 8 Guerra, J. (1989). Rapid serial enzyme measurements in
evaluation of patients with suspected myocardial infarction. American Joum_al_gf Cardiolggy,
§_3_, 652-655.

Eaker, E. D., (1989). Psychosocial factors in the epdemiology of coronary heart disease in
women. Pswchia_tric Clinics of North Amerig, 1_2_(1), 167-173.

Eisenberg, J. M. (1979). Sociologic influences on decision-making by clinicians. Annals of
Internal Medicine, @, 957-964.

Epstein, P. E. (1990). Cassandra and the clinician: Are clinical prediction rules changing the
practice of medicine? Annjatls oflntemiMedjcine. 113, 646-647.

Eysmann, A. B., & Douglas, P. S. (1992). Reperfusion and revascularization strategies for
coronary artery disease in women. M, fl, 1903-1907.

Fisher, M. L., Carliner, N. H., Becker, L. C., Peters, R. W., & Plotnick, G. D. (1983). Serum
creatine kinase in the diagnosis of acute myocardial infarction: Optimal sampling frequency.
M, 219, 393-394.

Gibler, W. B., Lewis, L. M., Erb, R. E., Makens, P. K, Kaplan, B. C., Vaughn, R. H., Biagini, A.
V., Blanton, J. D., & Campbell, W. B. (1990). Early detection of acute myocardial infarction in
patients presenting with chest pain and non-diagnostic ECGs: CK-MB sampling in the
emergency department. Annals of Emergency Medicine, fl, 1359-1365.

Goldman, L., Cook, E., Brand, D., Lee, T., Rouan, G., Weisberg, M., Acampara, D., Stasiulewicz,
C., Walshon, J., Terranova, G., Gottlieb, L., Kobemick, M., Goldstein-Wayne, B., Copen, D.,
Daley, K, Brandt, A, Jones, D., Mellors, J., & Jakubowski, R. (1988). A computer protocol to
predict myocardial infarction in emergency department patients with chest pain. Niw
England Journal of mm. 318, 797-803.

Greenland, P., Reicher-Reiss, H., Goldourt, U., & Behar, S. (1991). ln-hospital and 1-year
mortality in 1,524 women after myocardial infarction: Comparison with 4,315 men.

Circulation, Q, 484-491.

49

Greer, S., Dickerson, V., Schneiderman, L. J., Atkens, C., & Bass, R. (1986). Responses of male
and female physicians to medical complaints in male and female patients. Journal of Family
ELM. 23, 49-53.

Grimes, D. A. (1988). Prevention of coronary heart disease in women. America_i1 Joumw
Obstetricgnd Gynecology, 158, 1662-1688.

Hawthorne, M. A (1994). Gender differences in recovery after coronary artery surgery. Im_age_,
16, 75—80.

Hayward, R. A, McMahon, L. F ., 8. Bernard, A M. (1993). Evaluating the care of general
medicine inpatients: How good is implicit revievf? Anan of Lntemgl Medicine, m, 550-556.

Hedges, J. R., 8. Kobemick, M. S. (1988). Detection of myocardial ischemia/infarction in the
emergency department patient with Chest discomfort. Emerggncv Megm Clinics of North
Ame_ri_c_a_, _6_, 317-340.

Hedges, J. R., Rouan, G. W., Toltzis, R., Goldstein-Wayne, B., & Stein, E. A (1987). Use of
cardiac enzymes identifies patients with acute myocardial infarction otherwise unrecognized
in the emergency department. Annals of Emergengy Medicine, E, 249-252.

Herman, J. (1993). Gender differences in coronary artery disease: Bias, method or fact of life?
Archives of l:a_mily Medicine, g, 365-366.

Hibbard, J. H., 8. Pope, C. R. (1986). Another look at sex differences in the use of medical care:
Illness orientation and the types of morbidities for which services are used. Women & Hefialth,
11(2), 21 -36.

Horton, J. A. (1992). The women's health dgflook: A profile Of women's hea_lth in the United
§t_aE§. Washington, DC: Jacobs Institute of Women's Health.

lhlenfeld, J. T. (1988). Log-linear contigency table analysis: An illustration. In F. S. Downs
(Ed.), figmpook of reseJarch methodology (pp. 93-96). New York: American Journal of

Nursing.

50

Irvin, R. G., Cobb, F. R., 8 Roe, C. R. (1980). Acute myocardial infarction and MB creatine
phosphokinase: Relationship between onset of symptoms of infarction and appearance and
disappearance of enzyme. Archives of lntemgl Med_icine, 140, 329-334.

Jecker, N. S. (1991). Age-based rationing and women. M, £6, 3012-3015.

Joint Commission on Accreditation of Hospitals, (1983). Accreditation manual for hospitals,
Chicago: JCAH.

Kirschstein, R. L., 8 Merritt, D. H. (1985). Report of the Public Health Service task force on
women's health issues. Public Health Remus, 1m“ ), 73-106.

Lee, T. H., Cook, E. F., Weisberg, M., Sargent, R. K, Wilson, C., 8 Goldman, L. (1985). Acute
chest pain in the emergency room: Identification and examination of low risk patients.
Archives oflntemgl Med_icine, 145, 65-69.

Lee, T. H., 8 Goldman, L. (1986). Serum enzyme assays in the diagnosis of acute myocardial
infarction. Anna_ls of Ifltema1Medicine. 105. 221-233.

Lee, T. H., Rouan, G. W., Weisberg, M. C., Brand, D. A, Cook, E. F., Acampora, D., 8 Goldman,
L. (1987). Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24
hours of hospitalization. Anna_ls of l_ntema_l Medgne, 106, 181 -186.

Lerner, D. J., 8 Kannel, W. B. (1986). Patterns of coronary heart disease morbidity and mortality
in the sexes: A 26-year follow-up of the F ramingham population. America_n Heart Jogm_al,
m, 383-390.

Mair, J., Artner-Dworzak, E., Dienstl, A, Lechleitner, P., Morass, B., Smidt, J., Wager, I.,
Wettach, C., 8 Puschindorf, B. (1991). Early detection of acute myocardial infarction by
measurement of mass concentration of creatine kinase—MB. American Journal of Cardiolggy,
68, 1545-1550.

Marin, M. M., 8 Teichman, S. L. (1992). Use of rapid serial sampling of creatine kinase MB for
very early detection of myocardial infarction in patients with acute chest pain. American

Heart Joumgl, 123, 354-361.

51

Maynard, D., Althouse, R., Cerqueira, M., Olsuika, M., 8 Kennedy, J. W. (1991). Underutilization
of thrombolytic therapy in eligible women with acute myocardial infarction. American Journal
of Cardiology, 68, 529-530.

Neufeldt, V., 8 Sparks, A. N. (1990). lebster‘s newéworld dictionajj. New York: Warner
Books.

New York Heart Association (1979). Nomenclature and criteria for diagnosis of diseases of the
heart and great vessels (8th ed). Boston: Little, Brown.

Nomsis, M. J. (1992). SPSSIPC+ base system user's guide: Version 5.0. Chicago: SPSS Inc.

Plotnick, G. D., 8 Fisher, M. L. (1985). Risk stratification: Acost-effective approach to the

treatment of patients with chest pain. Archives of lntemal Medicine 115, 41-42.

 

Polit, D. F., 8 Hungler, B. (1991). Niising reasarch: Principles and methods (4th ed.)
Philadelphia, PA J. B. Lippincott Co.

Pozen, M. W., D'Agostino, R. 8., Mitchell, J. B., Rosenfeld, D. M., Gigielmino, B. A, Schwartz,
M. L., Tebagy, N., Valentine J. M., 8 Hood, W. B. (1980). The usefulness of a predictive
instrument to reduce inappropriate admissions to the coronary care unit. Annals of lntemal
Medicine, a2, 238-242.

Pozen, M. W., D'Agostino, R. B., Selker, H. P., Sytkowski, P. A, 8 Hood, W. B. (1984).
Predictive instrument to improved coronary-care-unit admission practices in acute ischemic
heart disease. New England Joum_al of Med_iLne, 310, 1273-1278.

Roberts, R. (1984). The two out of three criteria for the diagnosis of infarction: Is it passe?
Qpaat, Q, 511-513.

Roter, D., Lipkin, M., 8 Korsgaard, A. (1991). Sex differences in patients' and physicians'
communication during primary care medical visits. Medical Care, as, 1083-1093.

Schmidt, S. B., 8 Borsch, M. A. (1990). The prehospital phase of acute myocardial infarction in
the era of thrombolytics. Americain JournJal of Car—diologv. Q, 1411-1415.

Schroeder, J. S., Lamb, l. H., 8 Hu, M. (1978). The prehospital course of patients with chest

pain. American Journal of Medicine, 6_4, 742-747.

52

Tierney, W. M., Fitzgerald, J., McHenry, R., Roth, B. J., Psaty, B., Stump, D. L., 8 Anderson, F.
K. (1986). Physicians estimates of myocardial infarction in emergency room patients with
chest pain. Medigfl Decision MM, 6, 12-17.

Tobin, J. N., Wasserthal-Smoller, S., Wexler, J. P., Steingart, R. M., Budner, N., Lense, L., 8
Wachspreas, J. (1987). Sex Bias in considering coronary bypass surgery. Annals of
lntemalrMedicine, 107, 19-25.

Tofler, G. H., Stone, P. H., Muller, J. S., Willich, S. M, Davis, V. G., Poole, K, Strauss, H. H.,
Willerson, J. T., Jaffe, A. S., Robertson, T., Passamani, E., 8 Braunwald, E. (1987). Effects
of gender and race on prognosis after myocardial infarction: Adverse prognosis for women,
particularly black women. Journal of the Amerign College of Cardiolpgy, a, 473-482.

Underhill, S. L., Woods, S. L., Sivarajan Froelicher, E. S., 8 Halpenny, C. J. (1989). Cardfl
paraipg (2nd ed.) Philadelphia: J. B. Lippincott.

U. S. Department of Health and Human Services, Public Health Service (1990, September).
@lthypeople 2000: National health promotion and disezfl prevention objactives.

Verbrugge, L. M., 8 Steiner, R. R. (1981). Physician treatment of men and women patients: Sex
bias or appropriate care? Medical Care, fl, 609-632.

Wenger, N. K. (1990). Gender, coronary artery disease, and coronary bypass surgery. _Amala
of Interaaj Medicine, 112, 557-558.

World Health Organization (1959). Hypertension and coronary heart disease. Classification and

criteria for epidemiological studies. World Health Organization Technical, Series. 168, 3.

53

   

"Illli'llllllllliilllllflit?