Secondary Logo

Journal Logo

Effect of Incongruence of Acute Myocardial Infarction Symptoms on the Decision to Seek Treatment in a Rural Population

Morgan, Donna M. PhD, RN, CCRN

The Journal of Cardiovascular Nursing: September-October 2005 - Volume 20 - Issue 5 - p 365–371
ARTICLE
Free

People are advised to obtain immediate treatment for acute myocardial infarction (AMI), yet a delay occurs between the onset of symptoms and the decision to seek treatment. The objective of the study was to determine the extent of incongruence between expected and actual symptoms of AMI, effect of incongruence on decision time to seek treatment, and predictive effect of selected variables on decision time in a rural population. Ninety-eight rural patients receiving inpatient treatment for AMI at 2 hospitals in the Northeast of the United States from August 2001 through October 2002 completed the Morgan Incongruency of Heart Attack Symptoms Index and the Response to Symptoms Questionnaire. The median decision time was 93 minutes. There were differences between men and women regarding the symptoms that were expected to occur and that actually occurred. The extent to which symptoms interfered with the ability to carry out normal activities, degree of anxiety, and type of insurance coverage explained 23% of the variance in decision time. Although the expected symptoms did not match the actual symptoms, this incongruence did not affect decision time.

Donna M. Morgan, PhD, RN, CCRN Staff Nurse, Decker School of Nursing, Binghamton University, Binghamton, NY. Dr Morgan is now with Staff Nurse Skilled Nursing Facility, St Joseph's Hospital, Elmira, NY.

Corresponding author Donna M. Morgan, PhD, RN, CCRN, Skilled Nursing Facility, St Joseph's Hospital, 555 E Market St, Elmira, NY 14901 (e-mail: donnammorgan@direcway.com).

The author thanks Debra K. Moser, DNSc, RN, FAAN, for her comments on an earlier draft of this manuscript.

The incidence of acute myocardial infarction (AMI) in the United States in 2000 was 1,100,000, and 192,898 of these individuals died.1 Symptoms suggestive of AMI require immediate medical treatment to reduce morbidity and mortality. The public is advised to obtain treatment at once, yet for many persons there continues to be a considerable delay between the onset of symptoms and the decision to seek treatment. Persons likely to delay in seeking treatment demonstrated the following characteristics: older age, female, family member as a witness, and failure to recognize symptoms as cardiac in origin.2 There is a sparse amount of research specific to the decision time component of the delay to seek treatment for AMI, the major part of delay.3 Most of the research does not identify the rural/urban variable, or has been conducted in urban settings. Therefore, rural persons are underrepresented in the research.

A factor that may influence decision time is incongruence between expected and actual symptoms of AMI. Previously, investigators have demonstrated that many lay people have an unrealistic picture of symptoms that can be expected during an AMI.4 However, to date, no one has investigated the effect of incongruence between expected and actual symptoms on decision time. An analysis was conducted to determine whether, in a population of rural patients who have experienced an AMI, the degree of incongruence between expected AMI symptoms and the actual symptoms that occurred had an effect on decision time. The relationship between the extent of incongruence of symptoms and decision time, and the predictive effect of demographic, clinical, cognitive, emotional, and social factors upon decision time, was also investigated.

Back to Top | Article Outline

Background

No unified definition for rural exists. The US Census Bureau designates counties as urban or rural on the basis of metropolitan statistical area classification, a measure of economic and social integration with a large population nucleus.5 Rural persons experience barriers that influence their ability to receive healthcare, including lack of a public transportation system, limited choice of local providers, distance, lower income, and lack of health insurance.6 Living in a county without an acute care facility or living more than 2 miles from an established public transportation route are indicators of isolation and distance, key concepts in understanding rural health needs.7 Rural persons have characteristics that are different from those of their urban counterparts. They usually assess health as being able to perform one's work, with health per se being of secondary importance. They also tend to be more self-reliant than their urban counterparts, and this may influence their decision to seek assistance from others when experiencing symptoms suggestive of AMI.8

Back to Top | Article Outline

Methods

Design

This was a correlational study. Data were collected from patients who had experienced an AMI, with a focus on determining factors that contributed to the time required to reach the decision to seek treatment.

Back to Top | Article Outline

Sample and Setting

One of 3 criteria needed to be met to establish rural residency: living in a county designated by the US Census Bureau as rural, living in a county without an acute care facility, or living more than 2 miles from an established public transportation route. Rural patients (N = 110) were recruited from 2 hospitals in the Northeast of the United States from August 2001 through October 2002. Power analysis indicated a sample size of 100 for a moderate effect size, a power of 0.80, and α of .05. Patients were eligible for inclusion if they were currently receiving inpatient treatment for AMI, their symptoms began out of the hospital, and they had control over the decision to seek treatment.

Institutional review boards of the sponsoring university and clinical sites approved the study protocol. Following approval of the attending physician and informed consent from the patient, data collection took place in the patient's hospital room. Interviews were conducted within 3 days of admission for 60% of the patients, between day 4 and 9 for 39%, and on day 22 for 1 patient with a complicated admission. Most of the patients preferred to have the questions read to them, with answers recorded by the researcher and confirmed by the patients. Decision time interval, that portion of total time to treatment over which the patient had control, was defined as the time between symptom onset and the decision to seek medical help. To calculate decision time interval, the patient was asked the time of symptom onset and the time the decision was made to seek treatment or to go to the hospital. More commonly studied is time to treatment, the time from symptom onset to hospital arrival, and this includes transportation time.

Back to Top | Article Outline

Instruments

Morgan Incongruence of Heart Attack Symptoms Index

The Morgan Incongruence of Heart Attack Symptoms Index (MIHASI) was developed for this study to quantify the degree of incongruence between expected and actual symptoms of a heart attack. The first part of the MIHASI is a single question: “Overall, how well did the symptoms of your heart attack match what you expected a heart attack would be like?” Response options are as follows: (1) My symptoms matched completely what I expected. (2) Most of my symptoms matched what I expected. (3) Some of my symptoms matched what I expected, some did not. (4) Most of my symptoms did not match what I expected. (5) My symptoms did not match at all. A higher score indicates a greater degree of incongruence.

The second part is a list of 20 symptoms/descriptors of heart attack to which the patient responds twice: symptoms expected to occur with a heart attack and symptoms that actually occurred. Doctoral students in a nursing program assessed content validity of the individual items. Scoring involves assigning the value of “1” for each item that was marked “Expected to have” and not marked “Actually had,” or that was not marked “Expected to have” and marked “Actually had.” A score of “0” is assigned for each item that was marked both “Expected to have” and “Actually had,” or was not marked for either option. This scoring yields an incongruence value for each item, with 1 indicating incongruence and 0 indicating congruence. Adding the values of the individual items derives the total incongruence index, with a range from 0 to 20, with a higher score indicating greater incongruence. A master's-prepared RN with critical care expertise judged the instrument to have content validity for assessing the concept of incongruence.

Back to Top | Article Outline

Response to Symptoms Questionnaire

The Response to Symptoms Questionnaire (RSQ) was developed to assist in distinguishing between early and late responders to symptoms of AMI.9 Reilly and colleagues10 added one item to identify which, if any, family member was present at the time of symptom onset, and further additions were made to determine cognitive, symptom-appraisal, and social factors affecting the decision to seek treatment for AMI.11 The revised RSQ had an intraclass correlation greater than .85 for all items (P < .05) (D. K. Moser, DNSc, written communication, May 1, 2000). Two items added for the present study sought to identify the extent to which the symptoms interfered with the person's ability to carry out normal activities and the person's prior knowledge of angioplasty.

Back to Top | Article Outline

Statistical Analyses

Descriptive statistics were obtained for all variables. Sex comparisons were made using Pearson χ2 and independent t tests. Because of severely positively skewed decision times, square root transformations were performed and used for all calculations rather than the actual minutes, with medians reported to improve understanding of results. Spearman's ρ correlations were performed between decision time and ratio-level variables. Analysis of variance was used to examine the differences between groups of nominal-level independent variables and decision time. All analyses were carried out for both the total sample and by sex. Multiple regression analysis was used to analyze the relationship between the dependent variable of decision time and several independent variables. Alpha was set at less than or equal to .05 for all analyses.

Back to Top | Article Outline

Findings

Complete data were available for 98 of 110 patients interviewed. The sample included 62 men and 36 women, with a mean age of 61 and 65 years, respectively. There were no significant differences between males and females regarding living arrangements, employment status, education, history of coronary artery disease, angina, hypertension, heart failure, diabetes, or heart attack. Median decision time was 93 minutes for the total sample and men and 108 minutes for women.

Back to Top | Article Outline

Morgan Incongruence of Heart Attack Symptoms Index

The actual symptoms experienced by men and women are displayed in Table 1, with women experiencing significantly more fatigue than did men. Table 2 presents the percentages of expected and actual symptoms for the total sample. The 17 patients who did not experience pain, pressure, or discomfort in the chest had a median decision time of 60 minutes versus 95 minutes for those who had chest pain (t = 0.047, df = 96, P = .963). Responses to “Overall, how well did the symptoms of your heart attack match what you expected a heart attack would be like?” were as follows: “My symptoms completely matched what I expected,” 6%; “Most of my symptoms matched what I expected,” 18%; “Some of my symptoms matched and some did not match what I expected,” 28%; “Most of my symptoms did not match what I expected,” 16%; and “My symptoms did not match at all what I expected,” 30%. Spearman's ρ correlation between “Overall, how well did the symptoms of your heart attack match what you expected a heart attack would be like?” and decision time was significant only for men (r = 0.268, P = .037).

TABLE 1

TABLE 1

TABLE 2

TABLE 2

Mean total incongruence index was 7.31 for total sample, 7.29 for men, and 7.33 for women. Spearman's ρ correlation between “Overall, how well did the symptoms of your heart attack match what you expected a heart attack would be like?” and total incongruence index was significant for the total sample (r = 0.233, P = .022) and for men (r = 0.278, P = .030), but not for women. Spearman's ρ correlation between the total incongruence index and decision time was not significant. Men with a greater total incongruence index had a longer decision time, whereas women had a shorter decision time.

Back to Top | Article Outline

Response to Symptoms Questionnaire

Women (42%) were more likely to be alone when symptoms began than were men (16%) (χ2 = 7.817, df = 2, P = .020). The other person (family, nonfamily, or none) present at the onset of symptoms did not impact decision time. The most frequent first action taken by the first non-medical person when told about the symptoms was a suggestion to get medical help (33%); 12% suggested rest or medication, 10% contacted emergency medical services, and 10% transported the patient to the hospital. The first action taken by the other person present did not have a significant effect on decision time. The most frequent first action taken by the patient at onset of symptoms was an attempt to relax (24%); 21% took medication and 9% tried self-help remedy. Only 2% called emergency medical services as their first action. Patients' first action did not have a significant effect on decision time.

Stomach, hiatal hernia, or gall bladder was most often believed to be the source of the symptoms (31%), followed by heart (19%) and muscle pain (19%). Other sources included fatigue, flu, lung problem, no idea, high blood sugar, hard work, food poisoning, and arthritis. Source of symptoms did not have a significant effect on decision time.

Median decision time was 30 minutes for the 56% whose symptoms completely interfered with doing normal activities, 360 minutes for the 30% whose symptoms interfered somewhat, and 333 minutes for the 14% whose symptoms did not interfere at all (F = 8.586, P = .000).

Seriousness of the symptoms when first experienced was rated on a scale of 1 to 5, with 1 indicating “not at all” and 5 indicating “extremely serious.” The mean seriousness was 2.69, and those who rated their symptoms extremely serious had a shorter decision time (F = 2.549, P = .044). Level of discomfort or pain did not significantly affect decision time.

Anxiety was rated on a scale of 1 to 5, with 1 indicating “not at all” and 5 indicating “extremely anxious.” The mean anxiety score for the total sample was 2.5, with women more anxious than men ( for women = 2.85, for men = 2.30, t = −1.993, df = 95, P = .049). Decision time was significantly shorter for the total sample and men who reported a greater extent of anxiety, but not for women (F = 3.567, P = .009; F = 2.625, P = .044; F = 1.488, P = .230, respectively). Ability to control symptoms was rated on a scale of 1 to 5, with 1 indicating “none at all” and 5 indicating “an extreme amount of control.” The mean was 2.9, and those with more control over their symptoms had a significantly longer decision time (F = 3.703, P = .008).

Options for insurance coverage were uninsured, private insurance, or Medicare/Medicaid. Because of the age of the sample, many patients had both Medicare and private insurance, and this option was added for statistical analyses. Medicare and private insurance was held by 42% of the total sample, private insurance was held by 38%, Medicare/Medicaid was held by 14%, and 6% were uninsured. Seventeen percent of women were uninsured versus no men (χ2 = 11.007, df = 1, P = .001). Type of insurance coverage did not have a significant effect on decision time.

Sixty-one percent of patients had not heard about “clot busters” prior to their heart attack. Only 2 patients did not know about angioplasty. Knowledge of clot busters and angioplasty did not have a significant effect on decision time.

Back to Top | Article Outline

Predictors of Decision Time

Multiple regression was used to address the question “In a population of rural patients who have experienced an AMI, what is the predictive effect of demographic, clinical, cognitive, emotional, and social factors upon decision time?” Independent variables were entered into the model in sequence according to decreasing R2 values. The extent to which symptoms interfered with the ability to carry out normal activities had the highest R2 at 0.119, followed by extent of anxiety at 0.113, ability to control symptoms at 0.092, perceived seriousness of symptoms at 0.076, person with at onset of symptoms at 0.016, and type of insurance coverage at 0.016.

The 3 independent variables, extent to which symptoms interfered with the ability to carry out normal activities, degree of anxiety, and type of insurance coverage, resulted in the best model and explained 23% of the variance in decision time. Symptoms that completely prevented carrying out normal activities, symptoms that produced an extreme level of anxiety, and having Medicare and private insurance were likely to result in a shorter decision time.

Back to Top | Article Outline

Discussion

Thirty percent of the patients experienced symptoms that did not match at all what they expected a heart attack would be like. However, this lack of incongruence resulted in a longer decision time only for men, not for women. Zerwic et al reported longer delay times for patients with dissimilar symptoms.12 Johnson and King earlier reported a longer delay in receiving treatment for the 74% of patients who experienced a mismatch in symptoms.13

Incongruence measured by the total incongruence index was 7.31 out of a maximum of 20. This indicates there was a greater match than mismatch between expected and actual symptoms, and this contradicts the results reported in the preceding paragraph. So, which of these 2 methods, the single question or the incongruence index, provides a true reflection of the incongruence between expected and actual symptoms? One way to address this question is through analyzing the correlation between the 2 methods. Although the total incongruence index was much lower than the mismatch identified by the single question, a moderate positive correlation did exist (r = 0.278, P = .05). Some of the patients were unable to answer the single question initially and responded after identifying the individual symptoms expected and experienced. This alteration in sequence in answering the questions may have skewed the results.

Demographic factors that may have influenced decision time included sex, age, and insurance coverage. Men and women had similar decision times, and their ages were similar to the ages of participants in previous reports.12 Patients who had both Medicare and private insurance had the shortest decision time. This may be a reflection of a probable increased familiarity with the healthcare system, caused by older persons tending to have more chronic conditions and feeling more comfortable seeking healthcare. The effect of the type of insurance on decision time was not significant.

Clinical factors that affect decision time are the actual symptoms experienced and the effect of those symptoms on normal activities. Fifty-six percent had symptoms that completely prevented normal activities, with a decision time of 30 minutes, compared with a decision time of 333 minutes for the 14% who reported no interference (F = 8.586, P = .000). Treatment was usually sought when symptoms persisted despite self-treatment, or when intensity of symptoms increased. Women experienced significantly more fatigue than did men, consistent with the findings of Zuzelo.14

Significantly more patients expected to have, than actually had, shortness of breath, tiredness or fatigue, numbness in arm or hand, dizziness, sweating, or lightheadedness or feeling faint. It appears that these atypical symptoms of AMI were well known to the patients, though waiting for these symptoms to occur would likely add to the decision time. Symptoms were expected to be more severe than were actually experienced, with some patients referring to the media's tendency to portray AMI as a dramatic event.

The percentage (96%) of patients who expected to have chest pain was much higher than the percentage (71%) reported by Johnson and King.13 Chest pain was experienced by 83% of the patients compared with 53% reported by Gupta et al.15 The longer decision time by the 17 patients without chest pain was not significant. The present results conflict with the findings by Horne and colleagues,16 where patients without chest pain had a longer delay time than did those with chest pain (mean 11.5 hours vs 4.8 hours, t = 2.2, df = 87, P = < .05). Caution is needed when comparing these results, owing to the fact that Horne studied total delay time, not just decision time.

The median pain score was 8, an indication of the severity of pain, and does not coincide with 43% of patients stating symptoms were mild. Perhaps the patients were expecting the symptoms to be extremely severe. Severity of symptoms did not affect decision time, though it has been negatively correlated both with delay in seeking treatment17 and with having no effect on delay time.9 Severity of pain may be related to the extent to which the symptoms interfered with carrying out normal activities, which was a significant factor for decreasing decision time among women.

The perceived source of the symptoms is a cognitive factor that may affect decision time. Twenty patients attributed symptoms to their heart, and they had a decision time of 30 minutes. Recognizing symptoms as cardiac in origin has been reported to result in a significantly shorter delay time.11 Stomach, hiatal hernia, or gallbladder was selected most often, with a decision time of 150 minutes. Nausea or vomiting and diarrhea were experienced more often than anticipated, and its presence likely contributed to the frequent selection of stomach, hiatal hernia, or gallbladder as the source of symptoms. Education is needed to alert the public to the fact that gastrointestinal symptoms may indeed be caused by AMI. The longest decision time, 300 minutes, was for those who attributed their symptoms to a muscle problem. Decision time differences based on cause of the symptoms were not significant.

Emotional variables in decision time include anxiety and a feeling of control over the symptoms. Decision times based on extent of anxiety decreased significantly from 330 minutes for “not at all anxious” to 10 minutes for “extremely anxious.” An increased feeling of control over the symptoms did not negatively affect delay time in the research carried out by Reilly and colleagues.10 No consistent pattern emerged regarding the perceived ability to have control over the symptoms and may have been the result of difficulty in understanding the question.

The other person present when symptoms began and the first action taken by that person are social variables affecting decision time. The presence of a nonfamily member resulted in a shorter delay compared with a family member being present,10 while being with family, friends, others, or alone was not a predictor of delay in seeking treatment.18 Slightly more patients (67%) were with a family member than the 46%–61% reported by Dracup and colleagues,19 and the presence of a family member did not influence decision time. The most frequently reported first action taken by the other person present was a suggestion to get medical help. This would likely lead to a shorter decision time, though the suggestion to rest or take medication would prolong the decision time while the patient waits to assess the effect of the action. The first action taken by the other person did not significantly influence decision time.

Trying to relax or taking medication were the most frequently reported first actions taken by the patients, and both of these actions would seem to extend decision time while the patient waits to evaluate the effects of the intervention. Patients who ignored their symptoms had the longest decision times. Common sense would predict a longer time when symptoms are ignored, but there was no significant association between first action taken and decision time. Multiple regression indicated that symptoms that completely prevented carrying out normal activities, symptoms that produced an extreme level of anxiety, and having Medicare and private insurance accounted for 23% of the variance in decision time.

Back to Top | Article Outline

Limitations

Patients were recruited from 2 acute care facilities in the Northeast and may not be representative of all rural persons. Use of the US Census Bureau definition of rural counties resulted in several patients living near population centers of more than 50,000 people, and they may not have considered themselves to be rural. Another limitation was lack of reliability testing of the MIHASI, the instrument that quantified the extent of incongruence of symptoms. This study looked specifically at decision time, which is a part of the total delay time, and this limited the ability to make comparisons with other studies that generally use total delay time.

Back to Top | Article Outline

Conclusions

The vast majority of the variables discussed in this study cannot be changed, and so it becomes a matter of alerting the public to the reactions they can anticipate will occur in response to symptoms of AMI. Symptoms that interfere with the ability to carry out normal activities is a primary indication it is time to seek treatment. The urgency for seeking care becomes even greater when a high level of anxiety also exists. A major focus on heart attack education needs to be the responses the person may have to the symptoms, the feelings and actions rather than just the physical symptoms. It is also very important to let people know that their actual symptom experience may not match what they would expect to occur. Symptoms that are persistent, symptoms that interfere with normal activities, and a high level of anxiety are all very valid reasons to go to the emergency department, and it is not necessary to know the cause of the symptoms, or even suspect a heart attack. This information needs to be included in materials used for educating both the public and healthcare professionals, such as cardiopulmonary resuscitation instructional books and American Heart Association posters and booklets.

This study is the first known to the author to specifically focus on the AMI symptom experience in the rural population, an underrepresented group in the research literature. More research is needed in this population because of the key concepts in rural health of work beliefs and health beliefs, isolation and distance, self-reliance, lack of anonymity, outsider/insider, and old timer/newcomer. It also focused on the greatest component of delay to treatment, that being decision time.

AMI remains the leading cause of death for both men and women in the United States. The morbidity and mortality associated with AMI may be reduced as a result of a decrease in delay in seeking treatment when more is known about how people interpret and respond to their symptoms.

Back to Top | Article Outline

REFERENCES

1. American Heart Association. Heart Disease and Stroke Statistics—2003 Update. Dallas, Tex: American Heart Association; 2002.
2. Dracup K, Moser DK, Eisenberg M, Meischke H, Alonzo AA, Braslow A. Causes of delay in seeking treatment for heart attack symptoms. Soc Sci Med. 1995;40:379–392.
3. Zerwic JJ. Patient delay in seeking treatment for acute myocardial infarction symptoms. J Cardiovasc Nurs. 1999;13:21–32.
4. Zerwic JJ. Symptoms of acute myocardial infarction: expectations of a community sample. Heart Lung. 1998;27:75–81.
5. Rural Policy Research Institute. Metropolitan and non-metropolitan counties. 1999. Available at:
6. Curtiss CP. Trends and issues for cancer care in rural communities. Nurs Clin North Am. 1993;28:241–251.
7. Long KA, Weinert C. Rural nursing: developing the theory base. In: Lee HJ, ed. Conceptual Basis for Rural Nursing. New York, NY: Springer; 1998:3–18.
8. Bushy A. Rural determinants in family health: considerations for community nurses. In: Bushy A, ed. Rural Nursing. Vol 2. Newbury Park, Calif: Sage; 1991:133–145.
9. Mantell SJ, Berrios ED, Flanagan CH, et al. Psychological and social factors associated with early treatment in acute myocardial infarction. Circulation. 1989;80:(suppl 2) II–390.
10. Reilly A, Dracup K, Dattolo J. Factors influencing prehospital delay in patients experiencing chest pain. Am J Crit Care. 1994;3:300–306.
11. Dracup K, McKinley SM, Moser DK. Australian patients' delay in response to heart attack symptoms. Med J Aust. 1997;166:233–236.
12. Zerwic JJ, Ryan CJ, DeVon HA, Drell MJ. Treatment seeking for acute myocardial infarction symptoms: differences in delay across sex and race. Nurs Res. 2003;52:159–167.
13. Johnson JA, King KB. Influence of expectations about symptoms on delay in seeking treatment during a myocardial infarction. Am J Crit Care. 1995;4:29–35.
14. Zuzelo PR. Sex and acute myocardial infarction symptoms. MEDSURG Nurs. 2002;11:126–136.
15. Gupta M, Tabas JA, Kohn MA. Presenting complaint among patients with myocardial infarction who present to an urban public hospital emergency department. Ann Emerg Med. 2002;40:180–186.
16. Horne R, James D, Petrie K, Weinman J, Vincent R. Patients' interpretation of symptoms as a cause of delay in reaching hospital during acute myocardial infarction. Heart. 2000;83:388–393.
17. Hartford M, Karlson BW, Sjšlin M, Holmberg S, Herlitz J. Symptoms, thoughts, and environmental factors in suspected acute myocardial infarction. Heart Lung. 1993;22:64–70.
18. Burnett RE, Blumenthal JA, Mark DB, Leimberger JD, Califf RM. Distinguishing between early and late responders to symptoms of acute myocardial infarction. Am J Cardiol. 1995;75:1019–1022.
19. Dracup K, Moser DK, McKinley S, et al. An international perspective on the time to treatment for acute myocardial infarction. J Nurs Scholar. 2003;35:317–323.
Keywords:

AMI symptoms; decision time

© 2005 Lippincott Williams & Wilkins, Inc.