Friedman, Maureen M. PhD, RN; Quinn, Jill R. PhD, RN
Heart failure (HF) is a leading cause of hospitalization in the United States, especially among patients older than 65 years. Hospital discharges for HF have increased by 175% from 1979 to 2004. The estimated cost of HF in the United States for 2007 is $32.2 billion.1 Much of this cost is a result of numerous hospitalizations associated with HF. Some readmissions for HF are unnecessary because individuals with this condition wait a prolonged period, ranging from a few days to a number of weeks to seek care after symptom onset.2-5 Individuals delay contacting a healthcare provider until the intensity of the symptoms worsens and emergency department (ED) treatment is necessary. Recognizing symptoms of new or worsening HF may be difficult for patients, and deciding when to seek healthcare about a change in symptoms is even more challenging and difficult.
The most common symptoms of HF reported by patients in previous studies have been dyspnea/shortness of breath (SOB) and fatigue/tiredness2-9; these symptoms often worsen gradually over time rather than suddenly. Symptoms also can be subtle and fluctuate (eg, from bad to better to worse, and so on) during different times of the day or vary from day to day. Heart failure patients may not recognize these subtle changes and fluctuations in symptoms as important enough to report to their healthcare providers, which often contribute to treatment-seeking delays. It may be that patients also delay seeking treatment because they are not prepared to respond to early symptoms or changes indicative of decompensating HF. However, there is little evidence to support this possibility.
Counterintuitively, patients with chronic HF use diverse strategies to try to manage their HF-related symptoms before contacting their healthcare provider.10 These strategies include self-medicating, changing levels of physical activity, waiting a few minutes to perform an activity, completing activities at a slower pace, resting between activities, or staying quiet and dimming lights. Although these strategies are often taught to help patients manage their symptoms, they could add to the delay in seeking healthcare if the signs and symptoms are not recognized as potentially threatening changes that indicate worsening HF.
National guidelines clearly delineate the signs and symptoms of decompensating HF and provide healthcare providers with guidelines to promote self-care behaviors among HF patients, which include recognizing and managing their symptoms.11 Self-monitoring for symptom changes and deciding to seek appropriate care are difficult behaviors for patients with HF because of the multiple symptoms that they may experience and the fluctuations of the symptoms. In addition, the treatment regimen is complex.
Patient education is a key factor in promoting self-care behaviors. However, patient education alone does not result in effective behavior change for self-care and adherence.12-14 Artinian and colleagues12 found that 3 of the top 5 most frequently performed self-care behaviors were related to taking medications, and the 5 least frequently performed self-care behaviors involved HF symptom monitoring or management. Carlson and colleagues13 completed a survey of 139 HF patients and found that most participants had multiple HF symptoms during the past year, but their knowledge of the importance of signs and symptoms was poor; several participants had misperceptions related to self-care.
When symptoms of HF develop or worsen, adults with HF often report the symptoms to a family member, who helps them consider the options for action.15 The actual recommendations that the family members make to adults with HF who experience changes in symptoms are generally unknown. In addition, the actions that HF patients take, including whom they call before hospitalization, have not been explored. It is also unknown if persons with new onset HF experience symptoms differently and respond differently compared to those with a history of HF. Therefore, the purposes of this study were to (1) compare the presenting symptoms and treatment delay time of patients admitted for their first HF admission with that of HF patients with previous histories of hospital admission for HF, (2) examine whether there were differences for the 2 groups on whom they notified and when and what advice they received after the onset of their symptoms, and (3) describe and examine the differences in the actions taken by the patients in the 2 groups before their hospital admission. The study is unique in that it provides data directly from the perspective of hospitalized HF patients and compares 2 important groups of patients: those with a known history of HF and those without a previous history of HF. This perspective has not been studied previously in studies that explored delay in seeking care for HF.
Design, Sample, and Setting
A correlational design was used for this study. Participants were 50 years of age or older, mentally alert and able to complete the interview, without myocardial infarction or surgery on the current admission, hospitalized with a principal diagnosis of HF, and English speaking. Of the 247 eligible participants at the 3 acute care institutions in western New York, 212 (86%) consented to participate in the study; the remaining 35 (14%) patients did not feel up to completing the required interview during their hospitalization. Those who refused to participate were significantly older than those who agreed to participate (78.6 vs 72.5 years, t = −4.46, P < .001) and included more females (52.9%) compared to the participant group (48.6%) (χ2 = 5.2, P < .05).
Institutional review board approval was obtained from each of the 3 acute care institutions in western New York before the start of the study. Study participants were recruited by research assistants from the population of consecutive admissions for HF at each institution. After participants gave written informed consent to participate in the study, the research assistants, who were trained to use a standardized interview procedure, conducted a semistructured face-to-face interview. On average, the interviews were conducted on the third day after the participants' hospital admission. Clinical data were abstracted from the participants' medical records.
The Preadmission Illness Behavior Questionnaire is a semistructured questionnaire developed for this study. The Preadmission Illness Behavior Questionnaire measures the participants' symptom experiences and actions including their behaviors in seeking healthcare. The questionnaire was adapted for HF participants from Johnson and King's16 questionnaire for examining symptoms and symptom management strategies used by patients who had a myocardial infarction. Two advanced practice cardiovascular nurses confirmed that the items in the revised questionnaire measured the most common HF symptoms experienced by patients and the actions they used before hospital admission. Participants were asked open-ended questions about when their symptoms started before their hospital admission; whether the intensity of their symptoms changed at any point before their hospitalization; if yes, which symptoms changed in intensity and when; whom they contacted about their symptoms; the suggestions made by those who were contacted; and what actions they took while they were at home with symptoms. The instrument took 20 to 30 minutes to administer to participants.
The symptom checklist was developed for this study to capture the number and types of symptoms experienced by participants before hospitalization. The 13 items included on the checklist were based on the symptoms listed in the HF treatment guidelines developed by the American Heart Association.17 Content validity was established by 2 advanced practice cardiovascular nurses, who reviewed the checklist and confirmed that the 13 symptoms were the items most frequently reported by hospitalized HF patients at the time of hospital admission. Participants were asked if they had experienced each of the 13 symptoms in the 2 weeks before their admission, and the responses were recorded for each symptom as either 0 for no or 1 for yes. The responses to each symptom and the total number of symptoms were tallied for each subject. The instrument took approximately 10 minutes to administer. Cronbach α for the scale of 13 items was.68 for this study, indicating that the items had some independence from one another.
Participants' day and time of arrival in the ED were extracted from the participants' ED records. Cardiac and noncardiac comorbidities were obtained from the participants' admission history. Sociodemographic data (age, race, marital status, and years of education) were gathered directly from the participants.
Descriptive statistics was used to tabulate the participants' demographic data, cardiac histories, comorbidities, reported symptoms, the amount of time participants were home with symptoms until their arrival in the ED, the amount of time until family was called, the amount of time until the physician was called, the advice that the family members gave, the advice that the physicians gave, and the actions that the participants initiated. χ Tests and t tests were used to compare the group with an HF history with the group with no HF history on their demographic and clinical characteristics, symptoms reported, the advice from family members and physicians, and the participants' actions.
Because time with symptoms, intense time with symptoms, time until family was called, and time until the physician was called were positively skewed with broad ranges of time, a logarithmic transformation was used before further statistical analysis was performed. However, interpretation of results was based on raw hour data for clinical application, and times were reported using median values. t Tests were performed to compare the log-transformed means of all 4 time-related variables for the group with HF history and the group with no HF history.
The demographic and clinical data on the group with an HF history and the group with no HF history are presented in Table 1. The sociodemographic characteristics were similar for the 2 groups of patients. The most frequent underlying cardiac conditions for both groups of participants were coronary artery disease and hypertension. The only statistically significant difference in the clinical histories of the 2 groups was their history of documented cardiomyopathy. Diabetes was the most frequent noncardiac comorbidity documented in the medical history for both groups. None of the noncardiac comorbidities including chronic obstructive pulmonary disease, renal disease, arthritis, connective tissue disorders, cerebrovascular accident, and ulcers were significantly different for the 2 groups.
The symptoms experienced before hospitalization by the participants are presented in Table 2. Fifty percent or more of the participants in both groups reported experiencing the following symptoms on the symptom checklist: SOB on exertion, fatigue, orthopnea, weakness, edema, and cough. A significantly greater number of participants in the group with an HF history experienced fatigue compared to the group with no HF history (87.8% vs 76.6%, χ2 = 4.32, P < .05). The mean (SD) total number of symptoms experienced by participants with an HF history was slightly higher but not statistically different from the mean (SD) total number of symptoms for the group with no HF history (7.64 [2.4] vs 7.00 [2.7], P > .05). The median number of days that participants were home with symptoms before coming to the ED was 7 days for the 2 groups. Approximately half of the participants in both groups reported that one or more of their symptoms became more intense before their hospitalization; 52.8% of these patients said that SOB with exertion was the symptom that worsened over time. The median number of hours that participants with an HF history were at home with intense symptoms was 5.5 hours, whereas the group with no HF history had intense symptoms a median of 7.5 hours before their arrival at the ED. The log-transformed time with intense symptoms for the 2 groups was not statistically significant (t = −0.56, P > .05).
Most participants (89.2%) in both groups talked to a family member about how they were feeling before they came to the hospital, with many talking to more than one family member. The group with no HF history called a family member a median of 6 hours after their symptoms started, whereas the group with an HF history called a family member a median of 1.5 days after the symptoms started. The log-transformed time until a family member was called for the 2 groups was not statistically significant (t = .08, P > .05).
Of those participants who were married, 77% reported that they told their spouses about the symptoms they were having, whereas 55% of the unmarried participants told one or more of their children about their symptoms. The family members for the group with a history of HF versus the group with no HF history most often advised the participants to either call 911 (51.9% vs 38.5%, χ2 = 2.8, P > .05) or call their doctor (38.6% vs 47.4%, χ2 = 1.8, P > .05); only a small percentage of the family members advised the participants to wait and relax (5.3% vs 7.9%, χ2 = 0.19, P > .05) or do nothing (6.8% vs 8.8%, χ2 = 0.19, P > .05).
Fifty-nine percent of all the participants reported that they called their physician's office before they went to the ED. For the group with no HF history, the physician was called a median of 1.5 days after the symptoms started compared to a median of 3 days after symptoms started for the group with an HF history. The log-transformed time until a physician was called for the 2 groups was not statistically significant (t = .001, P > .05).
The advice given by physicians when called by the subject is summarized in Table 3. The most frequently advised action that the physicians gave to the participants was to call 911. The physicians advised the participants with an HF history to call 911 significantly more often than they advised those with no HF history to call 911. Physicians advised the participants with no HF history to come to the office (either on the day they called or wait for an upcoming appointment) significantly more often than they made those recommendations to participants with an HF history. A small percentage of participants in each group reported that their physician advised them by telephone to change one of their medications; the group differences were not statistically significant.
The actions that participants with an HF history took at home before admission were compared to the actions that participants without an HF history took (see Table 4). The percentage of patients with an HF history who reported that they called 911 was double the percentage of those with no HF history who called 911 (25% vs 12.5%). Only a small number of participants in both groups reported that they tried taking medications or resting to relieve symptoms before they came to the ED. However, those who tried taking medications in the 2 groups varied on the types of medications that they took. For the group with an HF history who tried taking medications (n = 31), the medication that 33% of that group tried was a diuretic. None of the participants without an HF history (n = 17) tried taking a diuretic to relieve their symptoms. This difference on the use of diuretics by the 2 groups was statistically significant (χ2 = 6.93, P < .001). The group with no HF history took a variety of medications other than diuretics, including cough medication, pain medication, antacids, and nitroglycerine to try to alleviate their symptoms. Approximately one-fifth of the participants in both groups reported that they took no particular action to try to alleviate their symptoms before they came to the ED.
The median time of approximately 7 days at home with symptoms before the arrival at the ED for both groups in this study is consistent with a previous research that indicates HF patients' delay in seeking care for new or worsening HF symptoms.3,4 Other investigators have found a wider time range of 3 days to a few weeks from the onset of symptoms until hospital admission.2,5 However, none of these studies compared participants with an HF history to those with no HF history before their admission.
The finding that the participants with an HF history were home with symptoms a similar amount of time as those with no HF history is a new finding that has not been described previously. This implies that patients with a previous history of HF do not appear to benefit from their previous experience with HF and do not seek healthcare more quickly for new or worsening symptoms than do those with no HF history. It could be that the education they received about symptoms of HF was different from what they actually experienced or that they adapted to their symptoms or tried to self-treat their symptoms. For those who try self-treating with medications, it may be that they wait to see if their symptoms improve as a result of their self-treatments before they call their healthcare providers and go to the ED for treatment.
Consistent with previous studies,3,7 HF patients seeking treatment in EDs present with a constellation of approximately 7 HF-related symptoms. The most common symptoms reported by the current sample were SOB and fatigue, which is consistent with previous research.2,4-6,8,9 A unique finding in this study was that a greater percentage of the participants with an HF history reported having fatigue compared to those without an HF history. This may be related to the fact that their HF condition was a more long-standing problem. Because fatigue may linger as a chronic symptom, patients with a history of HF may not view this symptom as noteworthy and may associate many of their other symptoms with lingering fatigue. The additional symptoms of orthopnea, weakness, edema, and cough experienced by 50% or more of the participants in both groups in this study were similar to some but not all of the symptoms reported by HF patients in previous studies.2-4,9
In this study, it appears that family members are called for advice before the physicians and before HF patients decide to go to the ED for treatment. Family members of persons with HF may be an important source of what House and colleagues18 called informational support or advice because they are the first to be notified and the first to offer advice on what the HF patient should do. The influence that this advice has on time to seek appropriate care by HF patients has been previously given little attention.
It is known from other research about HF patients who have memory and concentration difficulties that family members assist with symptom monitoring and healthcare decision making.10 From this study, it appears that family members play a pivotal role in decision making among patients presenting with new or worsening HF symptoms who do not have these same cognitive limitations. The family member recommended that the participants who were reporting HF symptoms should obtain healthcare by calling 911 or calling the physician. Family members, at least in this study, encouraged patients to seek some type of medical care. If symptoms could be detected before the symptoms became acute, then the choice of calling the physician and obtaining outpatient care at the physician's office could be more appropriately and safely recommended. Interventions for HF patients that include family members could be the key to reducing delay time to seek healthcare for HF symptoms.
The physicians were most often called after the family member was called. As reported by the participants, the physicians were more likely to advise the person with an HF history to call 911, whereas for those with no HF history, the physician frequently advised the person to come to the physicians' office. What is not known from these data is whether the group with an HF history actually reported more HF-related symptoms or greater severity to the physicians by telephone than did the group with no HF history. It is also not known whether the participants actually spoke to their physicians directly.
The difference in recommendations cannot be explained by differences in noncardiac comorbidities for the 2 groups because the 2 groups were similar on their noncardiac comorbidities. The group with an HF history was more likely to have a documented history of cardiomyopathy and a previous hospitalization for HF compared to the group with no HF history. It may be that the symptoms of the patients in the group with a previous history were interpreted as more serious by their physicians than that of the group with no history of HF. The participants' actual actions in both groups followed the reported recommendations from their physicians, with the group with an HF history calling 911 as their most frequently advised action and the group with no HF history obtaining an appointment at the physician's office as recommended.
One limitation of this study was the reliance on the mental recall by participants about their symptom experience for the days and sometimes weeks before hospital admission. The number of days, the order that symptoms occurred, and the nature of the directions given by family members and physicians are most likely a general picture of what the participants perceived had occurred before their hospital admission. In the future, investigators should consider verifying from family members and the physicians what advice they gave to HF patients to more fully describe the symptom management and decision making that HF patients used. In addition, examining what instructions providers gave to their patients and how well those instructions were understood would be helpful to determine the effectiveness of HF education in future research.
The standard measure of HF classification, the New York Heart Association HF classification, was not recorded in the records of many of the participants in this study. This meant that the researchers were not able to determine the symptom severity of the HF for participants with that index. Participants did have left ventricular ejection fraction recorded, and this value was used as a measure of HF severity.
The sample in this study was relatively small and was limited to those patients who were admitted to 3 acute care healthcare institutions in western New York state. Generalization may not extend beyond the type of participants hospitalized at the particular institutions sampled. Those who agreed to participate were younger and more likely to be male compared to those who refused to participate. This means that the findings cannot be generalized beyond these demographics of HF patients.
Summary and Clinical Implications
* The first action that both groups of HF participants initiated was to call a family member within 6 hours to 1.5 days of the symptom onset. Most family members of both groups gave similar advice: call the healthcare provider or call 911.
* The participants who had no HF history delayed contacting their healthcare providers about their new HF symptoms for a median of 1.5 days, whereas those with an HF history delayed contacting their healthcare providers for their worsening HF symptoms for a median of 3 days after the onset of symptoms. Both groups went on to delay their arrival at the ED for a median of 7 days after the onset of their HF symptoms.
* Both participants with an HF history and those with no HF history reported common symptoms of HF, including increasing SOB on exertion, fatigue, orthopnea, weakness, edema, and cough, that began a few to several days before admission to the ED. More participants with an HF history reported experiencing fatigue than those with no HF history.
* Physician-providers varied significantly on the advice they gave to the 2 groups of participants. The group with an HF history was more frequently advised by physicians to call 911 than the group with no HF history. The group with no HF history was more frequently advised by physicians to come to the office or wait for an upcoming appointment than the group with an HF history.
* The participants followed the actions that the healthcare provider suggested. This meant that those with an HF history called 911 more frequently than did the group with no HF history. The group with no HF history went to the healthcare provider's office or waited for an upcoming appointment. Ultimately, participants in both groups in this study went to the ED and were hospitalized.
One implication of these findings is that patients at high risk for the initial development of HF and decompensation of existing HF and their family members need to recognize a constellation of HF symptoms and act quickly when new or worsening symptoms occur. These symptoms should include the 10 to 15 symptoms outlined by the American Heart Association's HF treatment guidelines.11 Patients with an HF history should know that symptoms such as progressive fatigue may be a sign of worsening HF and should report this symptom to their healthcare providers as soon as fatigue increases. Those with an HF history also need to know that although they may have been instructed to adjust medications such as diuretics as needed for fluid retention and reduce activities to minimize symptoms, those attempts at self-treatment should be time limited so that contact with the provider is not excessively delayed.
Appropriate giving of advice by family members should be reinforced by care providers in the future because a family member appears to play a key role as the first person notified about symptoms. Because the patients did not act quickly on the advice given by the family members to call the physician or call 911, the family may need to more emphatically stress their advice to their loved one with new or worsening HF symptoms to seek healthcare quickly.
Because the most effective interventions to reduce hospital readmission rates for HF patients in previous studies have included close home monitoring by nurses and other healthcare professionals,19-23 it may be that in the future, HF patients and family caregivers need more direct assistance for better symptom recognition and prompt action to seek nonemergent care. New and less costly methods of assisting HF patients, including those at risk for developing HF, with self-monitoring of symptoms and decision making for care seeking need to be developed and tested. This assistance could aid patients and their families to recognize changes in HF symptoms sooner and promptly alert providers.
1. American Heart Association (AHA). Heart Disease and Stroke Statistics: 2007 Update
2. Schiff GD, Fung S, Speroff T, McNutt RA. Decompensated heart failure: symptoms, patterns of onset, and contributing factors. Am J Med
3. Friedman MM. Older adults' symptoms and their duration before hospitalization for heart failure. Heart Lung
4. Jurgens CY. Somatic awareness, uncertainty, and delay in care-seeking in acute heart failure. Res Nurs Health
5. Evangelista LS, Dracup K, Doering LV. Treatment-seeking delays in heart failure patients. J Heart Lung Transplant
6. Mayou R, Blackwood R, Byrant B, Garnham J. Cardiac failure: symptoms and functional status. J Psychosom Res
7. Parshall MB, Welsh JD, Brockopp DY, Heiser RM, Schooler MP, Cassidy KB. Dyspnea duration, distress, and intensity in emergency department visits for heart failure. Heart Lung
8. Stanek EJ, Oates MB, McGhan WF, Denofrio D, Loh E. Preferences for treatment outcomes in patients with heart failure: symptoms versus survival. J Card Fail
9. Todero CM, LaFramboise LM, Zimmerman LM. Symptom status and quality-of-life outcomes of home-based disease management program for heart failure patients. Outcomes Manag
10. Bennett SJ, Cordes DK, Westmoreland G, Castro R, Donnelly E. Self-care strategies for symptom management in patients with chronic heart failure. Nurs Res
11. American College of Cardiology/American Heart Association (ACC/AHA). Heart Failure Guidelines
12. Artinian NT, Magnan M, Christian W, Lange MP. What do patients know about their heart failure? Appl Nurs Res
13. Carlson B, Riegel B, Moser DK. Self-care abilities of patients with heart failure. Heart Lung
14. Ni H, Nauman D, Burgess D, Wise K, Crispell K, Hershberger RE. Factors influencing knowledge of and adherence to self-care among patients with heart failure. Arch Intern Med
15. Friedman MM. Social support sources among older women with heart failure: continuity versus loss over time. Res Nurs Health
16. Johnson J, King K. Influence of expectations about symptoms on delay in seeking treatment during myocardial infarction. Am J Crit Care
17. Konstam M, Dracup K, Baker D, et al. Clinical practice guideline no 11: heart failure: evaluation and care of patients with left-ventricular systolic dysfunction. Rockville, MD: Agency for Health Care Policy and Research, Public Health Services, US Depart of Health and Human Services; 1994, AHCPR Publication No: 94-0612.
18. House JS, Umberson D, Landis KR. Structures and processes of social support. Annu Rev Sociol
19. Hanumanthu S, Butler J, Chomsky D, Davis S, Wilson JR. Effect of a heart failure program on hospitalization frequency and exercise tolerance. Circulation
20. Naylor MD, Brooten DA, Campbell RL, Maislin G, McCauley KM, Schwartz JS. Transitional care of older adults hospitalized with heart failure: a randomized, control trial. JAGS
21. Rich MW, Beckham V, Wittenberg C, Leven CL, Freedland KE, Carney RM. A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure. NEJM
22. Simon S, Horowitz JD. Home-based intervention in congestive heart failure: Long-term implications on readmission and survival. Circulation
23. West JA, Miller NH, Parker KM, et al. A comprehensive management system for heart failure improves clinical outcomes and reduces medical resource utilization. AJ Cardiology
© 2008 Lippincott Williams & Wilkins, Inc.