The American Journal of Medicine
Volume 121, Issue 2 , Pages 127-135, February 2008

Physician Attitudes Toward End-Stage Heart Failure: A National Survey

Preliminary data presented at the 9th annual meeting of the Heart Failure Society of America, Boca Raton, Florida, September 2005.

  • Paul J. Hauptman, MD

      Affiliations

    • Division of Cardiology, Saint Louis University School of Medicine, St. Louis, Mo
    • Center for Outcomes Research, Saint Louis University School of Medicine, St. Louis, Mo
    • Dr. Hauptman was funded by National Institutes of Health (NIH) grant RO1 AG 021515.
    • Corresponding Author InformationRequests for reprints should be addressed to Paul J. Hauptman, MD, Division of Cardiology FDT-15, Saint Louis University Hospital, 3635 Vista Avenue, St. Louis, MO 63110.
    • ,
  • Jason Swindle, MPH

      Affiliations

    • Center for Outcomes Research, Saint Louis University School of Medicine, St. Louis, Mo
    • ,
  • Zainal Hussain, MD

      Affiliations

    • Department of Medicine, Saint Louis University School of Medicine, St. Louis, Mo
    • ,
  • Lois Biener, PhD

      Affiliations

    • Center for Survey Research, University of Massachusetts, Boston.
    • ,
  • Thomas E. Burroughs, PhD

      Affiliations

    • Center for Outcomes Research, Saint Louis University School of Medicine, St. Louis, Mo

Article Outline

Abstract 

Background

Despite recent improvements in medical therapies, heart failure remains a prevalent condition that places significant burdens on providers, patients, and families. However, there is a paucity of data published describing physician beliefs about heart failure management, especially in its advanced stages.

Methods

In order to better understand physician decision-making in end-stage heart failure, we used a stratified random sampling of physicians obtained from the Master File of the American Medical Association to survey cardiologists (n=600), geriatricians (n=250), and internists/family practitioners (n=600).

Results

Response rate was 59.6% (highest among geriatricians). The vast majority (>90%) of respondents cited similarities between the clinical trajectory of end-stage heart failure and lung cancer or chronic obstructive pulmonary disease; however, only 15.7% stated that they could predict death at 6 months “most of the time” or “always.” Inpatient volume was a predictor of confidence in predicting mortality (odds ratio=1.38, 95% confidence interval, 1.36-1.40). Less than one quarter of respondents formally measure quality of life. The experience with deactivation of implantable cardioverter defibrillators was limited: 59.8% of cardiologists, 88.0% of geriatricians, and 95.1% of internal medicine/family practice physicians have had 2 or fewer conversations with patients and families about this option.

Conclusions

Significant gaps in knowledge about and experience with end-stage heart failure exist among a large proportion of physicians. The growing prevalence and highly symptomatic nature of heart failure highlight the need to further evaluate and improve the way in which care is delivered to patients dying from the disease.

Keywords: Heart failure, Survey

 

Heart failure is a prevalent, progressive, and potentially lethal condition1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13 that places burdens on providers, patients, and patients’ families. Although epidemiological studies differ according to diagnostic criteria and sampling methods, they provide similar estimates of high prevalence and low survival rates once advanced disease has developed.7, 14, 15, 16, 17, 18 The impact on the health care system is enormous. Annual hospitalizations with a primary diagnosis of heart failure increased in absolute terms by 294,000 from 1985 to 199519 and to a total of over 1 million in 2005.12 The number of patients over the age of 65 years admitted with a first-listed diagnosis of heart failure increased from 20.3 to 22.1 per 1000 Medicare enrollees from 1990 to 2000.10 The actual numbers may be even higher as International Classification of Diseases-9 coding may undercount patients admitted with clinical evidence for decompensated heart failure.20 With respect to outpatient care, heart failure has been ranked as the 10th most common reason for an office visit in older patients.21

Clinical Significance

 


•Less than one quarter of physicians specializing in geriatrics, internal or family medicine, or cardiology believe that they can accurately predict death in heart failure patients within a 6-month time frame.

•This lack of confidence affects physician decision-making about use of intravenous inotropic drugs, deactivation of implantable cardioverter defibrillator devices, and referral to hospice care.

•Interventions to improve physician decision-making in end-stage heart failure are needed.

Despite this, there is a paucity of published data about physician beliefs and practices regarding end-stage heart failure. Research has demonstrated that important differences exist in diagnostic and therapeutic approaches according to physician specialty, which are partly explained by case mix.22, 23, 24, 25, 26, 27, 28 Among hospitalized heart failure patients, a small fraction (5%) has a “Do Not Resuscitate” order.29 Hospice is rarely provided at the time of hospital discharge,30, 31 and assessments of patient preferences for resuscitation are frequently inaccurate.32 Despite the increasing number of patients and the nature of the clinical challenge, there are few published studies on physician attitudes toward key issues in advanced heart failure care such as risk stratification, device management, and selection of care options.

To explore these issues, we designed a national survey of physicians. We evaluated the degree to which physicians believe that they can accurately predict death in a 6-month time frame, because one of the keys to clinical care in heart failure is the ability to reasonably predict the trajectory of disease. Uncertainty about a patient’s prognosis might impact a number of key decisions, including timing of hospice referral and deactivation of implantable cardioverter defibrillators, and hence, a patient’s experience with their illness. Similarly, we hypothesized that only a minority of physicians would report use of quality-of-life measures despite the highly symptomatic nature of advanced heart failure and the utility of these measures to facilitate patient care. This issue has added resonance given the presence of options such as inotropic therapy that might shorten life span while improving quality of life.

Back to Article Outline

Methods 

Instrument Development 

Based on a literature review, information generated in focus groups, and cognitive interviews, we conducted a pilot study of internists, family practice physicians, geriatricians, and cardiologists (n=68, response rate 38%) selected at random from the American Medical Association Master File of Physicians (www.ama-assn.org/ama/pub/category/2673.html) to examine item response and several data collection strategies. Subsequently, a random stratified sample of 1450 physicians (cardiology: 600; internal medicine: 300; family practice: 300; and geriatrics: 250) was selected based on power calculations of the pilot data. To ensure the inclusion of physicians with experience prescribing inotropic drugs, 90 known prescribers identified by a previous study33 were specifically recruited; these responses were weighted in all subsequent analyses. Eligibility was determined according to several criteria: in active practice, board-certified or -eligible in the particular specialty, not a participant in research or administration more than 50% of the time, and direct involvement in the care of heart failure patients within the prior year.

In an effort to improve response rates before the mailing of surveys, press releases were sent out to physician newsletters. Initial surveys were sent with a cover letter on university letterhead, a second letter of support from an opinion leader in the different disciplines, and a 10-dollar cash incentive. If no response was received within 2 weeks, the survey was re-sent using priority overnight mail.

Response Rates and Sample Weighting 

We calculated the overall response rate using the methodology of Ayanian et al.34 Potential respondents were divided into categories: survey completed and returned (n=734); survey returned but respondent was ineligible (n=117); survey returned with note that physician was deceased (n=1); unable to trace during mailing or by phone (n=19); or follow-up phone call determined ineligibility (n=20). Respondents who returned a blank survey or declined were deemed potentially eligible. Applying the eligibility rate to the remaining nonresponders (n=498), the response rate was calculated as the number of responses returned (n=734) divided by the eligible number (734+498) yielding a final response rate of 59.6%.

Because the sample design utilized stratification by physician specialty, sampling weights were calculated based on the original probabilities of selection from the American Medical Assocation database for members of each stratum. Separate response weights were calculated for the general sample and for the over-sampled inotropic prescribers. The individual weights were combined into a composite weight that was used in analyses in which the strata were combined.

Data Management 

Responses to survey questions were double data entered. Data files were checked for completeness and converted into SPSS (SPSS Inc., Chicago, Ill) and SAS (SAS Institute Inc., Cary, NC) for analysis. Duplicate returns (n=2) were deleted by determining if one response was more complete; otherwise one was chosen at random.

Data Analysis 

In comparing responders with nonresponders, we found lower response rates for male physicians (54.6% vs 62.0%, P=.025), cardiologists (51.4% vs 64.8% for geriatricians and 57.7% for internal medicine/family practice physicians, P=.002), and physicians with a graduation date before 1970 (50.4% vs 55.6% for 1970-1979, 53.3% for 1980-1989, and 62.7% for 1990-2000s, P=.021). There were no significant differences in response rates by geographic region.

For the substantive questions relating to knowledge and practice, we examined the patterns of response according to practice specialty, year of graduation, physician sex, and patient volume. We examined the distribution of each variable to make certain that the distributional requirements for subsequent data analysis were met. Descriptive statistics were calculated for each of the key physician characteristics and interview questions. We used chi-squared and Student’s t tests to test for significant differences across the physician specialties.

Multivariate logistic regression models were developed to test the relative impact of key factors on physician self-report of ability to accurately predict 6-month mortality and physician self-report of the frequency of discussions with patients/family members about implantable cardioverter defibrillator deactivation. The .05 significance level was set as the criterion for entry and removal of covariates from the model at each step. The Wald chi-squared test was used to evaluate statistical significance. We report odds ratios (OR) with 95% confidence intervals (CI). Candidate predictor variables for the models were based on factors that were likely to impact physician behavior, including volume of heart failure patients, sex, physician specialty, formal training in palliative care, use of clinical guidelines in practice, year of training, practice type and setting, and presence of a cardiac catheterization laboratory or palliative care team at the physician’s primary hospital.

The study was approved by the Institutional Review Boards of the respective institutions. The funding source had no role in the design, conduct, analysis, or reporting of the study.

Back to Article Outline

Results 

Characteristics of Physician Responders 

Physician characteristics are provided in Table 1. A total of 734 respondents were available for analysis, with representation from cardiologists (n=292), internists/family practice (n=291), and geriatricians (n=151); the majority were male (74%). The median year of medical school graduation was 1980 for cardiologists, 1985 for geriatricians, and 1986 for internal medicine/family practice physicians.

Table 1. Physician Demographics and Practice Characteristics among Responders
Demographics and Practice CharacteristicsCardiologistsGeriatriciansFamily Practice/Internal MedicineP Value
Medical school graduation year (median)198019851986P>.10
Post-graduate medical training year (median)198719911990P>.10
n%n%n%
Primary practice type P<.0001
Staff HMO62.185.3175.8
Private practice with university affiliation5819.92113.93010.3
Private practice without university affiliation17058.23825.216055.0
University practice/faculty4716.12013.282.7
Veterans Administration31.0117.362.1
Nursing homena 1711.382.7
Community health centerna 117.33512.0
Solo vs group practice P<.0001
Solo3411.63321.98127.8
Group25386.611475.520369.8
Practice setting
Office/clinic based16857.58657.025788.3P<.001
Hospital based10034.2127.9175.8
Both office/clinic and hospital51.710.720.7
Practice location P<.0001
Urban13746.97046.48328.5
Suburban12944.24731.111840.5
Rural237.93019.98127.8
Number of outpatient HF patients seen per month P<.0001
None20.796.0124.1
1-5217.25335.112141.6
6-105017.13523.27726.5
11-156020.52315.23612.4
16-256221.21610.6289.6
26-404515.496.0113.8
>405017.142.641.4
Number of inpatient HF patients seen per month P<.0001
None62.13120.59532.6
1-53411.65435.812041.2
6-108228.13724.53913.4
11-157224.7117.3186.2
16-255518.8127.9124.1
26-40268.921.320.7
>40165.532.031.0
Formal hospice training237.99764.26823.4P<.0001
Involved in clinical research13947.62013.2186.2
Percentage of patients with traditional fee-for-service Medicare P=.004
0%41.453.393.1
<10%124.142.6206.9
10-24%258.61711.34615.8
25-49%6020.51711.33913.4
50-75%8629.54026.57224.7
>75%9331.86643.79633.0
Percentage of patients with capitated managed care plan P=.05
0%5719.52919.27626.1
<10%7927.14630.57626.1
10-24%8127.73523.25117.5
25-49%4415.12013.24013.7
50-75%144.8106.6258.6
>75%72.453.3165.5
Primary hospital has a catheterization lab27293.211374.821072.2P<.0001
Primary hospital has palliative care specialists18061.69663.616556.7P<.21

Abbreviations: HMO=health maintenance organization; na=not applicable; HF=heart failure.

Sample sizes vary and not all percentages add up to 100% because of missing responses.

Most respondents were primarily office-based (cardiologists 60.0%, geriatricians 61.5%, and internal medicine/family practice physicians 90.5%). Urban practice settings were most common except for internal medicine/family practice physicians who were more likely to be located in a suburban community. The predominant setting was a private group practice without a university affiliation. Cardiologists were more likely to work primarily in a facility with a cardiac catheterization laboratory than were geriatricians or internal medicine/family practice physicians (94% vs 76% and 73%, respectively), and to be involved in some type of clinical research (47% vs 13% and 6%, respectively). Cardiologists also reported a higher volume of outpatient and inpatient heart failure patients. More than half of physicians reported access to palliative care specialists (cardiologists 68%, geriatricians 67%, and internal medicine/family practice physicians 61%). However, cardiologists were more likely to report training in hospice, palliative care, or the death/dying process than were geriatricians and internal medicine/family practice physicians (59% vs 40% and 37%, respectively). Across all specialties, fee-for-service Medicare coverage was the most common insurance type.

Ability to Predict Disease Trajectory 

There was considerable variability in physician confidence about the ability to accurately predict clinical trajectory, in particular, patient death, within a 6-month time frame. Only 15.7% of all physicians reported that they could “always/most of the time” predict the timing of death, whereas the majority reported that they could either “rarely/never” (25.2%) or only “sometimes” (59.1%) predict death. Univariate analysis revealed that confidence in predicting 6-month mortality was higher among geriatricians than cardiologists or internal medicine/family practice physicians (P<.001). Other physician characteristics reaching significance included large clinical volume in the inpatient setting (P<.001), a history of formal training in palliative care (P=.008), and a stated willingness to objectively measure functional status (P=.001) and refer patients to hospice (P=.009). In contrast, confidence was not associated with university affiliation or year of medical school graduation (P>.10 for both). When examining these relationships multivariately, factors that reached significance included higher inpatient heart failure volumes (OR 1.38, 95% CI, 1.36-1.40), female sex (OR 1.70, 95% CI, 1.65-1.76) and group practice setting (OR 1.48, 95% CI, 1.44-1.54) (Table 2).

Table 2. Predictors of Confidence in Assessing Likelihood of 6 Month Mortality⁎
FactorOdds Ratio95% Wald
Inpatient HF volume1.3771.3591.395
Outpatient HF volume0.9790.9670.991
Sex
Male1.0——
Female1.7041.6461.763
Cardiologist vs FP/IM/Geriatrics
FP/IM/Geriatrics1.0——
Cardiologist0.8090.7710.849
Palliative care training
Yes1.0——
No0.9460.9120.982
Year of postgraduate training0.9850.9830.987
Practice setting
Urban1.0——
Suburban1.2261.181.273
Rural1.321.2681.375
Group vs solo practice
Solo1.0——
Group1.4851.4361.535
Palliative care team/individual at primary hospital
Yes1.0——
No0.7370.7140.761
Published guidelines help in decision-making
A great deal1.0——
A moderate amount0.9820.9091.062
A small amount0.7500.6960.807
Not at all0.4070.3770.439

Abbreviations: HF=heart failure; FP=family practice; IM=internal medicine.

⁎Confidence=response of “Always” or “Most of the Time” to the question: How often can you accurately predict the occurrence of death within 6 months in a patient with end-stage heart failure?

Physicians were asked to rate the importance of a number of potential predictors of mortality; New York Heart Association class (55.5%), findings on physical examination (48.0%), serum creatinine (47.1%), presence of pulmonary hypertension or abnormal hemodynamics (50.7%), ejection fraction (54.5%), and access to care (49.9%) were considered “very important.” When asked to select the most important factor, respondents most frequently rated ejection fraction (27%) and New York Heart Association class (19%). Among those expressing a high degree of confidence in determining the timing of mortality, the physical examination and weight loss were rated highest.

In terms of mode of death, the 2 most commonly selected were unexpected sudden death/arrhythmia and progressive pulmonary edema with dyspnea. The prognosis of heart failure was reported to resemble the prognosis of chronic obstructive lung disease or lung cancer rather than other chronic conditions (such as acquired immunodeficiency syndrome or amyotrophic lateral sclerosis). The clinical trajectory of heart failure was most commonly described as “intermittent declines in functional status with subsequent improvements but not to baseline,” rather than as “a continuous downhill course” or “intermittent declines with return near baseline for varying times” upon resolution of the exacerbation. In the terminal stages of the disease, dyspnea was most often characterized as moderate-to-extreme; pain was most frequently characterized as mild-to-moderate (Table 3).

Table 3. Descriptions of the Clinical Course of End-Stage Heart Failure
CardiologistsGeriatriciansFamily Practice/Internal MedicineP Value
n%n%n%
Clinical trajectory of progressive HF P<.01
A⁎72.564.131.1
B†5118.13524.06322.5
C‡15053.48256.217261.4
D§7326.02315.74215.0
Level of dyspnea in terminal stages of HF P<.0001
Extreme3010.63624.34415.4
Severe13246.66946.616156.5
Moderate11540.63624.37426.0
Mild62.174.762.1
None00.000.000.0
Level of pain in terminal stages of HF P<.01
Extreme10.310.700.0
Severe82.864.1124.2
Moderate6322.35436.510135.3
Mild14551.86443.212844.8
None6422.72315.54515.7
Prognosis of end-stage HF resembles P<.0001
Lung cancer13548.93322.57626.7
COPD11441.310873.519066.7
AIDS/ALS/IBD/MS279.864.1196.7

Abbreviations: HF=heart failure; COPD=chronic obstructive pulmonary disease; AIDS=acquired immune deficiency syndrome; ALS=amyotrophic lateral sclerosis; IBD=inflammatory bowel disease; MS=multiple sclerosis.

⁎A=

Continuous downhill course.

†B=

General downhill course in functional status with small short-lived improvements.

‡C=

Intermittent decline in functional status with subsequent improvements, but not to baseline, and overall worsening trajectory.

§D=

Intermittent declines in functional status with return near baseline for varying times.

Management Decisions 

Cardiologists were significantly more likely to consider implantable cardioverter defibrillator deactivation to be appropriate in terminal heart failure (cardiologists 87%, geriatricians 78%, internal medicine/family practice physicians 70%). However, discussions with patients and families about potential device deactivation are very uncommon across all physician specialties (Table 4). For example, 60.0% of geriatricians and 75.4% of internal medicine/family practice physicians reported that they had never talked with patient or family members about this option. Frequency of conversations about deactivation was associated with higher inpatient heart failure volumes (OR 1.49, 95% CI, 1.47-1.51), female sex (OR 1.21, 95% CI, 1.16-1.26), cardiology specialty (OR 5.07, 95% CI, 4.82-5.34), formal training in palliative care (OR 1.43, 95% CI, 1.38-1.50), use of guidelines in clinical decision-making (OR 3.22, 95% CI, 2.90-3.58), and group practice setting (OR 1.84, 95% CI, 1.77-1.92).

Table 4. Experience with Deactivation of Implantable Defibrillators
Device DeactivationCardiologistsGeriatriciansFamily Practice/Internal MedicineP Value
n%n%n%
Frequency of discussion with family P<.0001
Never6924.79060.021575.4
1-2 times9835.13926.05619.7
3-9 times8630.8149.3144.9
>10 times869.374.700.0
Frequency of discussion with primary care physician
Never10035.5
1-2 times11440.4
3-9 times5318.8
>10 times155.3
Frequency of discussion with cardiologist P<.001
Never 8254.719970.3
1-2 times 5134.06824.0
3-9 times 138.7155.3
>10 times 42.710.4
Appropriate to not replace an ICD generator in end-stage HF
Yes25691.1
No258.9

Abbreviations: ICD=implantable cardioverter-defibrillator; HF=heart failure.

Sample sizes vary due to missing responses.

The prescription of or experience with intravenous inotropic therapy is limited among noncardiologists (not used in past year: 68% geriatricians and 63% internal medicine/family practice physicians). In addition, although not supported by published data or clinical practice guidelines, some practitioners believe that this therapy prolongs life (3% cardiologists and 13.4% noncardiologists) and is cost effective (3% cardiologists and 12% noncardiologists). Consistent with these findings, most physicians would only rarely or sometimes select treatment options that alleviate symptoms but shorten length of life (cardiologists 81%, geriatricians 71%, internal medicine/family practice physicians 76%).

In general, physicians believe that they should be the providers who initiate discussions about death and dying rather than other health care or ancillary providers such as nurses or pastoral care staff. Further, cardiologists believe that cardiologists should initiate these discussions (73%), whereas internal medicine/family practice physicians (78%) and geriatricians (76%) largely believe that they are best positioned to initiate the discussions.

Experience with hospice referral for heart failure also appears to be uncommon. A significant percentage cited uncertainty about the timing of referral (cardiologist 64%, geriatricians 45%, internal medicine/family practice physician 68%) and the likelihood of patient acceptance (cardiologist 47%, geriatrician 56%, and internal medicine/family practice physician 44%) as factors that limited wider use of this option at or near the end of life.

Despite high recognition of heart failure symptoms, few physicians reported using formal quality-of-life instruments during outpatient visits (cardiologists 17%, geriatricians 22%, internal medicine/family practice physicians 22%) or conducting 6-minute corridor walk tests to assess functional status (cardiologists 19%, geriatricians 20%, internal medicine/family practice physicians 10%). Further, the majority of physicians stated that they only rarely or sometimes discuss the tradeoff between quality and quantity of life with the patient or patient’s family. The physician’s outpatient volume of heart failure patients did not predict response to these questions.

Physician Knowledge and Guidelines 

There was consensus about the utility of guidelines for heart failure; most physicians indicated that guidelines helped them a moderate (47%) or great amount (23%) in clinical decision-making. Despite this, 21% of respondents reported that angiotensin-converting enzyme inhibitors are contraindicated if the serum creatinine is >2.0 mg/dL; 27% mistakenly reported that intermittent outpatient inotropic therapy is considered a reasonable practice according to the guidelines of the American College of Cardiology; 33.9% underestimated heart failure prevalence in the United States; and 75.2% underestimated the 1-year mortality for Medicare patients hospitalized with heart failure.

Back to Article Outline

Discussion 

End-stage heart failure is increasingly common, as the prevalence of the disease increases and the population continues to age. Despite significant advances in medical therapy and initiatives to improve compliance with evidence-based approaches, mortality remains high. Further, the increased use of implantable defibrillators may be responsible for an increase in the population potentially at risk for dying from progressive pump dysfunction rather than sudden arrhythmic death.35, 36 Nevertheless, little is known about physicians’ perception of this patient population, the clinical approach that physicians use, and the biases underlying their choice of approach.

We found that there are considerable gaps in knowledge and confidence about the appropriateness of certain management options including use of intravenous inotropic drugs, referral to hospice, and implantable cardioverter defibrillator deactivation. Much of the difficulty appears to be related to uncertainty about the clinical trajectory and the likelihood of death within a 6-month period, a challenge that may not be shared in oncological disease.37 The variability in clinical status is significant in advanced heart failure and may complicate decision-making.38 Nevertheless, a number of physician characteristics were identified as associated with greater degrees of confidence, including the volume of heart failure patients under the care of the physician, a history of formal training in palliative care, and participation in a group practice setting.

Prior attempts at understanding physician practice in heart failure have utilized surveys22, 25, 39 and single23, 40, 41 and multi-center42, 43 observational studies. Differences in the demographics of the heart failure patient have been noted according to physician specialty. The degree to which these clinical experiences influence physician attitudes cannot be determined from the current survey; however, although differences among specialties exist, we observed many commonalities. For example, despite the increased burden of symptoms, most physicians do not formally measure quality of life nor objectively measure functional status; the former has been recently described.44 Further, descriptions of the trajectory of disease and the mode of death were similar across physician groups.

It is important to elucidate biases underlying treatment decisions in patients with end-stage heart failure. This is an essential step before designing interventions that might help physicians select appropriate care options and evaluate the proper timing for initiation of end-of-life discussions.45 For example, the perception that dyspnea is severe and highly prevalent at the end of life may not be accurate.46 It also is noteworthy that, despite the growth in use of device therapy, there is a general lack of experience with device deactivation, which almost certainly has direct implications for patients and patient families.47 Similarly, the failure to objectively evaluate quality of life or functional status suggests that physicians may not appropriately recognize the impact of heart failure on their patients. Therefore, although the precise elements in quality improvement or physician education initiatives in heart failure have not been established,48 our data suggest that the need is significant.

Limitations 

The degree to which physicians’ self-reports accurately describe practice is not firmly established; we are not able to determine how physician attitudes impact the selection of care options. However, in a previous study, Wennberg and colleagues found that self-reported propensities for cardiac catheterization and imaging stress tests were linked to population-based angiography rates.49 Our response rate (59.6%) is acceptable for a physician survey, but we had limited data available with which to compare responders with nonresponders. Further, we do not know the relative frequency with which the different specialties provide care to patients with end-stage heart failure and hence the direct implications of our findings, although estimates suggest that the majority of heart failure care is delivered by internists and family practice physicians rather than cardiologists.50

In summary, significant gaps in knowledge about and experience with decision-making in end-stage heart failure appear to exist among a large proportion of physicians. The growing prevalence of the disease and its highly symptomatic nature highlight the need to further evaluate and improve the way in which care is delivered to patients dying from heart failure.

Back to Article Outline

Acknowledgments 

The authors thank the staff of the Center for Survey Research at the University of Massachusetts. We acknowledge with thanks Drs. Robert Rakel, W.R. Hazzard, and Eugene Braunwald, who provided supporting letters that accompanied the surveys.

Back to Article Outline

References 

  1. Gillum RF. Epidemiology of heart failure in the United States. Am Heart J. 1993;126:1042–1047
  2. Graves EJ. Detailed diagnoses and procedures, National Hospital Discharge Survey 1991. Vital Health Stat 13. 1994;(115):1–290
  3. Brophy JM. Epidemiology of congestive heart failure: Canadian data from 1970 to 1989. Can J Cardiol. 1992;8(5):495–498
  4. Cowie MR, Mosterd A, Wood DA, et al. The epidemiology of heart failure. Eur Heart J. 1997;18(2):208–225
  5. Ghali JK, Cooper R, Ford E. Trends in hospitalization rates for heart failure in the United States, 1973-1986 (Evidence for increasing population prevalence). Arch Intern Med. 1990;150:769–773
  6. McDonagh TA, Morrison CE, Lawrence A, et al. Symptomatic and asymptomatic left-ventricular systolic dysfunction in an urban population. Lancet. 1997;350(9081):829–833
  7. Schocken DD, Arrieta MI, Leaverton PE. Prevalence and mortality rate of congestive heart failure in the United States. J Am Coll Cardiol. 1992;20:301–306
  8. Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive heart failure in the community: trends in incidence and survival in a 10-year period. Arch Intern Med. 1999;159(1):29–34
  9. Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive heart failure in the community: a study of all incident cases in Olmsted County, Minnesota, in 1991. Circulation. 1998;98(21):2282–2289
  10. Brown DW, Haldeman GA, Croft JB, et al. Racial or ethnic differences in hospitalizations for heart failure among elderly adults: Medicare, 1990 to 2000. Am Heart J. 2005;150:448–458
  11. Baker DW, Einstadter D, Thomas C, Cebul RD. Mortality trends for 23,505 Medicare patients hospitalized with heart failure in Northeast Ohio, 1991-1997. Am Heart J. 2003;146:258–264
  12. Rosamond W, Flegal K, Friday G, et al. Heart disease and stroke statistics—2007 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2007;115;:e69–e171
  13. Kosiborod M, Lichtman JH, Heidenreich PA, et al. National trends in outcomes among elderly patients with heart failure. Am J Med. 2006;119:616e.1–616e.7
  14. Cowie MR, Wood DA, Coats AJ, et al. Incidence and aetiology of heart failure: a population-based study. Eur Heart J. 1999;20(6):421–428
  15. Morgan S, Smith H, Simpson I, et al. Prevalence and clinical characteristics of left ventricular dysfunction among elderly patients in general practice setting: cross sectional survey. BMJ. 1999;318(7180):368–372
  16. Ho KKL, Pinsky JL, Kannel WB, Levy D. The epidemiology of heart failure: the Framingham Study. J Am Coll Cardiol. 1993;22(4 Suppl A):6A–13A
  17. Kupari M, Lindroos M, Iivanainen AM, et al. Congestive heart failure in old age: prevalence, mechanisms and 4-year prognosis in the Helsinki Ageing Study. J Intern Med. 1997;241(5):387–394
  18. Levy D, Kenchaiah S, Larson MG, et al. Long-term trends in the incidence of and survival with heart failure. N Engl J Med. 2002;347:1397–1402
  19. Haldeman GA, Croft JB, Giles WH, Rashidee A. Hospitalization of patients with heart failure: National Hospital Discharge Survey, 1985 to 1995. Am Heart J. 1999;137(2):352–360
  20. Goff DC, Pandey DK, Chan FA, et al. Congestive heart failure in the United States (Is there more than meets the I(CD Code)? The Corpus Christi Heart Project). Arch Intern Med. 2000;160:197–202
  21. Schappert SM. National Ambulatory Medical Care Survey: 1991 Summary. Vital Health Stat 13. 1994;(116):1–110
  22. Bello D, Shah NB, Edep ME, et al. Self-reported differences between cardiologists and heart failure specialists in the management of chronic heart failure. Am Heart J. 1999;138(1 Pt 1):100–107
  23. Reis SE, Holubkov R, Edmundowicz D, et al. Treatment of patients admitted to the hospital with congestive heart failure: specialty-related disparities in practice patterns and outcomes. J Am Coll Cardiol. 1997;30(3):733–738
  24. Chin MH, Wang JC, Zhang JX, Lang RM. Utilization and dosing of angiotensin-converting enzyme inhibitors for heart failure (Effect of physician specialty and patient characteristics). J Gen Intern Med. 1997;12(9):563–566
  25. Chin MH, Friedmann PD, Cassel CK, Lang RM. Differences in generalist and specialist physicians’ knowledge and use of angiotensin-converting enzyme inhibitors for congestive heart failure. J Gen Intern Med. 1997;12(9):523–530
  26. Fleg JL, Hinton PC, Lakatta EG, et al. Physician utilization of laboratory procedures to monitor outpatients with congestive heart failure. Arch Intern Med. 1989;149:393–396
  27. Auerbach AD, Hamel MB, Davis RB, et al. Resource use and survival of patients hospitalized with congestive heart failure; differences in care by specialty of the attending physician. Ann Intern Med. 2000;132:191–200
  28. Hltatky MA, Fleg JL, Hinton PC, et al. Physician practice in the management of congestive heart failure. J Am Coll Cardiol. 1986;8:966–970
  29. Wachter RM, Luce JM, Hearst N, Lo B. Decisions about resuscitation: inequities among patients with different disease but similar prognoses. Ann Intern Med. 1989;111:525–532
  30. Jaagosild P, Dawson NV, Thomas C, et al. Outcomes of acute exacerbation of severe congestive heart failure: quality of life, resource use, and survival. SUPPORT Investigators. The Study to Understand Prognosis and Preferences for Outcomes and Risks of Treatments. Arch Intern Med. 1998;158(10):1081–1089
  31. Hauptman PJ, Goodlin SJ, Lopatin R, et al. Characteristics of patients hospitalized with acute decompensated heart failure who are referred for hospice care. Arch Intern Med. 2007;167:1990–1997
  32. Krumholz HM, Phillips RS, Hamel MB, et al. Resuscitation preferences among patients with severe congestive heart failure: results from the SUPPORT project. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments. Circulation. 1998;98(7):648–655
  33. Hauptman PJ, Mikolajczak P, Mohr CJ, et al. Chronic continuous home inotropic therapy in end-stage heart failure. Am Heart J. 2006;152:1096.e1–1096.e8
  34. Ayanian J, Hauptman PJ, Guadagnoli E, et al. Belief and practices of generalist and specialist physicians regarding drug therapy for acute myocardial infarction. N Engl J Med. 1994;331:1136–1142
  35. Bardy GH, Lee KL, Mark DB, et al. Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT) Investigators Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med. 2005;352:225–237
  36. Lindenfeld J, Feldman AM, Saxon L, et al. Effects of cardiac resynchronization therapy with or without a defibrillator on survival and hospitalizations in patients with New York Heart Association Class IV heart failure. Circulation. 2007;115:204–212
  37. Teno JM, Weitzen S, Fennell ML, Mor V. Dying trajectory in the last year of life: does cancer trajectory fit other diseases?. J Palliat Med. 2001;4:457–464
  38. Hauptman PJ, Masoudi F, Weintraub W, et al. Variability in the clinical status of patients with advanced heart failure. J Card Fail. 2004;10:397–402
  39. Edep ME, Shah NB, Tateo IM, Massie BM. Differences between primary care physicians and cardiologists in management of congestive heart failure: relation to practice guidelines. J Am Coll Cardiol. 1997;30(2):518–526
  40. Chin MH, Wang JC, Zhang JX, et al. Differences among geriatricians, general internists, and cardiologists in the care of patients with heart failure: a cautionary tale of quality assessment. J Am Geriatr Soc. 1998;46(11):1349–1354
  41. Lowe J, Candlish P, Henry D, et al. Specialist or generalist care? (A study of the impact of a selective admitting policy for patients with cardiac failure). Int J Qual Health Care. 2000;12(4):339–345
  42. Ahmed A, Allman RM, Kiefe CI, et al. Association of consultation between generalists and cardiologists with quality and outcomes of heart failure care. Am Heart J. 2003;145:1086–1093
  43. Rutten FH, Grobbee DE, Hoes AW. Differences between general practitioners and cardiologists in diagnosis and management of heart failure: a survey in every-day practice. Eur J Heart Fail. 2003;5(3):337–344
  44. Kosiborod M, Soto GE, Jones PG, et al. Identifying heart failure patients at high risk for near-term cardiovascular events with serial health status assessments. Circulation. 2007;115:1975–1981
  45. Hauptman PJ, Havranek E. Integrating palliative care into heart failure care. Arch Intern Med. 2005;165:374–378
  46. Derfler MC, Jacob M, Wolf RE, et al. Mode of death from congestive heart failure: implications for clinical management. Am J Geriatr Cardiol. 2004;13:299–306
  47. Goldstein NE, Lampert R, Bradley E, et al. Management of implantable cardioverter defibrillators in end-of-life care. Ann Intern Med. 2004;141:835–838
  48. Abraham WT. Preceptorships: a practical approach to education in heart failure. J Card Fail. 1999;5:265–268
  49. Wennberg DE, Dickens JD, Biener L, et al. Do physicians do what they say? (The inclination to test and its association with coronary angiography rates). J Gen Intern Med. 1997;12:172–176
  50. O’Connell JB, Bristow M. Economic impact of heart failure in the United States: time for a different approach. J Heart Lung Transplant. 1994;13:S107–S112

PII: S0002-9343(07)00927-8

doi:10.1016/j.amjmed.2007.08.035

The American Journal of Medicine
Volume 121, Issue 2 , Pages 127-135, February 2008

Article Tools