eMJA     The Medical Journal of Australia

Home | Issues | eMJA shop | My account | Classifieds | Contact | More... | Topics | Search   

Hospitalisation for congestive heart failure: burden and outcomes

Fiona M Blyth, Ross Lazarus, David Ross, Michael Price, Gary Cheuk and Stephen R Leeder

Electronically published Monday 2 June 1997. Please submit comments by Monday 30 June 1997.


This article has been accepted for publication in the Medical Journal of Australia after undergoing peer review (you can read the reviews). It has been revised in response to the reviewers' comments, but has not yet been edited by the MJA. Readers are invited to comment on the article. These comments may be considered by the editors and authors in preparing the article for print. Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company
Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au/>".
  • CHF is an important cause of hospital admission and readmission in the elderly
  • Established CHF is associated with significant morbidity and mortality
  • Monitoring the burden of CHF through routine data sources is difficult

Abstract - Introduction - Methods - Results - Sample characteristics - Preadmission characteristics - Length of stay - Readmissions - Deaths - Changes in domicile - Quality of life - Patient knowledge - Discussion - Acknowledgements - References - Authors' details - Figure 1 - Figure 2 - Table 1 - Table 2 - Table 3 - Table 4 - Table 5 - Table 6


MJA reviewers' comments - Register to be notified of new articles by e-mail - To Papers for review list - ©MJA1997

 

Abstract

 Objective: To describe the hospital burden and health outcomes associated with admission for congestive heart failure (CHF).
Design: Descriptive follow-up study.
Setting: Tertiary level metropolitan teaching hospital.
Patients: Acute adult inpatients with a clinical diagnosis of CHF for a period exceeding 24 hours, admitted to Westmead Hospital, Sydney during the 4 months from September 1993 to January 1994. At baseline, 122 patients were assessed; 88 patients were assessed at 4 month follow-up.
Interventions: Usual clinical care.
Main outcome measures: Length of stay, hospital bed days, readmissions, mortality, health related quality of life (SF-36), patient knowledge.
Results: The average age of subjects was 73.4 years. Many were using informal domiciliary care before admission. During the baseline admission period, mean length of stay was 13.8 days accounting for 1683 hospital bed days or 4.2% of bed days for all inpatients aged 65 years and over. 15 patients were readmitted for CHF during the 4 month follow-up period, with a total of 26 CHF-related admissions. Twenty one patients (17.2%) died during the course of the study. Quality of life at baseline was poor compared with population normative data, with slight improvement among survivors at 4 month follow-up. Patient knowledge of CHF was poor in a subsample survey (n= 24).
Conclusions: Chronic CHF represents a significant burden to patients (through morbidity and mortality), their carers (through provision of daily care), and (through multiple admissions for acute decompensation). It is difficult to monitor the hospital burden of CHF using routine data sources.
©MJA 1997

 

Introduction

 Congestive heart failure (CHF) is an important clinical and public health problem. It is the fastest growing cardiovascular disorder in the USA1 and the only one increasing in incidence and prevalence.2 In Scotland CHF hospital discharge rates have risen in a decade to be almost identical to those for myocardial infarction.3

CHF has been estimated to affect 3%-5% of those aged over 65 years,4 and 10% of those over 75 years. It is the leading cause of hospital admission and hospital readmission in Americans aged over 65 years.5 In 1990, CHF cost the US economy $US8 billion, and accounted for 5 million hospital days.6,7 In the UK, the National Health Service spends £360 million per year in diagnosis and management, on a par with spending on stroke or asthma. Most of this expenditure is on hospital admissions.8

Major improvements in CHF morbidity and mortality have been demonstrated in trials of angiotensin converting enzyme (ACE) inhibitor drugs.4,9,10 However, when this mortality reduction is applied to the population taking these drugs, there is a mean increase in survival of less than six months.2

Little information is available that describes the outcomes of current hospital management of CHF patients.11,12 This paper describes the health outcomes of a cohort of patients hospitalised with CHF, and the impact of their hospitalisation on a health service.  

Methods

 All acute adult inpatient admissions to Westmead Hospital during a four month period (September 1993 to January 1994) either admitted with a clinical diagnosis of CHF, or who developed CHF of more than 24 hours duration during an admission, were eligible for inclusion in the study. In all cases, the diagnosis of CHF was made by medical staff on clinical grounds. Additional eligibility criteria for obtaining data by questionnaire were fluency in English and absence of significant cognitive impairment. Patients were identified by: review of the computerised admissions log; medical record chart audit in the relevant ward areas; and consultation with medical staff.

Two cardiology trained Research Nurses implemented an active, daily case-finding strategy, in order to overcome difficulties encountered in identifying CHF patients relying on existing record systems alone. Diagnostic codes are generally not available until some time after discharge. Moreover, a hospital admission for CHF may be assigned a Principal ICD-9 code which reflects the underlying cause of their CHF, in which case one of the secondary ICD-9 codes for that admission should contain a CHF-related code. It was found that the existence of such a secondary code in a medical record does not automatically identify instances of active CHF - a past medical history of CHF may also be coded with a secondary code.

Potential subjects were assessed for their general condition, fluency in English and where appropriate screened for cognitive impairment using the Mini-Mental State examination.13 Informed consent was then sought. If patients were unable to participate fully, consent was sought to review their medical record, and follow them up at four months to ascertain vital status.

Baseline data were collected by interview, medical record review and self-administered questionnaires. Data were collected on: demographic characteristics; domiciliary arrangements; current and past medical history; clinical severity of heart failure using the New York Heart Association (NYHA) criteria for grading functional incapacity of patients with cardiac disease; investigations; drug treatment; in-hospital resource use ; formal and informal use of domiciliary care; and health-related quality of life (HRQOL) before admission, measured with the SF-36 (MOS 36 Item Short Form Health Survey).14 Patient knowledge about CHF was assessed by a short questionnaire in a subset of 24 consecutive patients.

Four months after baseline admission, subjects were contacted by telephone to arrange follow-up by post. Data were sought on: health status; HRQOL; current treatment; and domiciliary arrangements. The hospital's computer system was searched for any readmissions during the follow-up period.

All analysis was performed using SAS version 6.08 for Windows15 Two sample t-tests were used to assess the statistical significance of differences between groups, with adjustments made for multiple comparisons. SF-36 data were scored and subscales were calculated using the recommended scoring algorithm.14 Baseline and follow-up SF-36 scores were compared using paired t tests.

Westmead Hospital Human Research Ethics committee approval was obtained for the study, and informed consent obtained from all study subjects.

 

Results

 There was no "gold standard" available for checking the accuracy and completeness of study case ascertainment methods. However, a list was assembled of all separations from Westmead Hospital during the baseline study period that were assigned a Principal Diagnosis code for CHF (428.0,428.1,428.9). These were then cross-checked against a list of study subjects. This list identified 7 patients who were not identified by study case-finding methods. These were considered "missed" potential cases, but represented a small proportion of this group (3.4%).

During the four month baseline recruitment period, 154 patients met the required clinical criteria, and 122 of this eligible patient group (79%) consented to participate in the study. In the remaining group of 32 patients who met the clinical criteria (21%), three declined to participate and the other 29 were unable to participate in the study for reasons relating to either their overall medical condition or their inability to communicate in English.

The final study sample included patients who, while willing to contribute, were not well enough to participate fully in the study. Rather than exclude these patients entirely, permission was sought to review their medical records and follow them up at four months for assessment of vital status. This group contained 37 of the 122 patients (30%) in our study cohort. The remaining 85 patients (70%) underwent full baseline data collection.

At four months, follow-up data were available from 58 (68%) of the 85 subjects with full baseline assessments. Most common reasons for failure to follow-up were: death before follow-up (n= 15, 17.7%) and extreme poor health (n= 6, 7.1%). Excluding deaths, those lost to follow-up tended to be older than the sample as a whole, more likely to be female (n= 9, 75%), and had relatively severe CHF (NYHA grade III at baseline, indicating marked limitation in physical activity). In the group who had medical record review only at baseline (n= 37) six deaths (16.2%) occurred before follow-up.  

Sample characteristics

 Females made up slightly more than half of the cohort (54.9%) and were older on average than males (76.8 compared with 69.4 years, P= 0.0001) (Table 1). Most subjects had a history of established CHF, which tended to be relatively severe (NYHA grade III or IV). Ischaemic heart disease (58.2%) and hypertension (39.3%) were major aetiological factors for CHF. Many of the sample had significant comorbidities. The cohort was cared for by cardiologists (n= 74, 60.7%), and geriatricians (n= 48, 39.3%).

Forty five subjects (36.9%) had heart-failure-specific admission diagnoses, and another 38 (31.2%) had suggestive diagnoses (Table 2). When admission diagnoses were compared with ICD-9 codings, 39 (86.6%) of those with heart-failure-specific admission diagnoses and 30 (78.9%) of those with suggestive diagnoses had a CHF-related ICD-9 code (428, 428.0, 428.1, 428.9) in one of the first six ICD-9 coding positions.
 

Preadmission characteristics

 Data on preadmission domiciliary circumstances were available for 99 subjects. Most ( n= 88, 72.2%) lived in private accommodation. Many of the cohort reported needing regular help from family or friends with such basic tasks as general housekeeping (n= 50, 50.5%), and supervision of medication (n= 23, 23.2%).  

Length of stay

 The mean length of stay during the baseline admission was 13.8 days (sd = 12.1). Length of stay had a markedly skewed distribution, with a median value of 10 days (range, 2-66). Overall, the sample accounted for 1683 bed days during baseline admissions. They accounted for 7.6% of hospital separations (excluding day-only patients) and 10.6% of hospital bed days associated with the geriatricians and cardiologists who participated in the study during the 4-month recruitment period. Of those patients aged 65 years and over admitted during the same period, the CHF cohort accounted for about 4.2% of bed days.  

Readmissions

 During the four month follow-up period, there were 73 readmissions to Westmead Hospital from the study cohort. About one third of these readmissions (n= 26, 35.6%) were due to a further episode of CHF, occurring in 15 subjects. A total of 171 bed-days were used for CHF-related readmissions by the 14 subjects for whom data were available.  

Deaths

 A total of 21 subjects died during the course of the study (17.2%).  

Changes in domicile

Eight (8.1%) were discharged to a higher level of domiciliary care, suggesting a decline in independence and increased use of health and/or community services.  

Quality of life

 Data on HRQOL before admission were obtained from 84 subjects at baseline (Table 3). Mean SF-36 subscale scores for these subjects were generally low, particularly for subscales with physical health components. For each subscale there was a wide range of scores, indicating substantial variability in health-related quality of life before admission. Compared with Australian normative SF-36 data for males and females aged 65 years and over,16 the study cohort reported significantly lower mean HRQOL for all subscales (Figure 2).

Follow-up SF-36 data were obtained for 58 subjects (Table 4). Subjects with missing follow-up data had either died before follow-up (n= 15) or were too unwell to fill in the form (n= 6). In that sense, the SF-36 results at follow-up represent the survivor' population within the study cohort. Mean follow-up subscale scores in survivors were somewhat higher compared with their baseline scores. The smallest improvements were in the subscales related to physical health. Improvements in the Social Function, Vitality and Mental Health subscales were statistically significant.
 

Patient knowledge

 The 24 subjects who completed the questionnaire were similar to the CHF cohort as a whole in regard to age, sex, length of stay, and type of treating specialist. Most (n= 19, 79.2%) had experienced a prior hospital admission for CHF. Less than half (n= 11, 45.8%) knew that they had been diagnosed as having heart failure. Although most agreed ( n= 16, 66.7%) that patients with heart failure would need to take medication permanently, 6 (25.0%) thought that it was unlikely to recur. Most patients agreed that shortness of breath (n= 19, 79.2%) and ankle swelling (n= 15, 62.5%) were important symptoms but few (n= 4, 16.7%) recognised rapidly increasing weight as important.  

Discussion

 CHF impacts significantly on hospital services. During the period that patients were recruited, the study cohort contributed significantly to adult bed days attributable to participating clinicians. A substantial proportion (12.3%) of the cohort were readmitted with CHF within 4 months of their baseline admission, suggesting an annual readmission rate of around 36%. This compares to an annual readmission rate of nearly 20% in the SOLVD registry study cohort who were younger than the Westmead cohort.12 CHF admissions show seasonal variations with a winter peak associated with chest infection. Therefore, the impact of CHF on acute hospital services may be even greater at that time of year.

The 48 subjects (39.3%) who were admitted under the care of geriatricians were, by hospital admission policy, those aged over 65 with serious comorbidity and/or likely to represent a placement problem post-discharge. Many subjects required substantial help with tasks essential for independent living, and most of this help was provided informally. This is clearly a group which would require increased resources with deteriorating health.

HRQOL was generally poor in this group compared with normative data,16 probably reflecting the combined effects of disease severity and comorbidity. There was limited improvement between baseline and follow-up, particularly in the subscales related to physical health, perhaps indicating that at baseline the lower limits of some SF-36 subscales did not adequately reflect the subjects' condition (i.e. a "floor" effect was operating). The mean changes in scores over time were small, but underlying this was great variability between individuals.

At the time this study was conducted, there were very few educational resources targeted at CHF patients. This was somewhat surprising, as CHF is a chronic condition with acute exacerbations which may be ameliorated by early recognition and intervention on key signs and symptoms and with a need for compliance with medication. It may be that for some CHF patients hospital admission could at times be avoided by timely intervention prompted by self-monitoring of signs and symptoms. 76 subjects (62.3%) had a recorded history of acute deterioration lasting for more than 24 hours, suggesting a possible window of opportunity for early recognition and intervention.

While the patient knowledge survey was limited in scope, it demonstrated an apparent lack of understanding of key features of the condition, remembering that most of the subsample had previous experience of hospitalisation for CHF. Recent acute illness could have contributed to this. The finding that almost one quarter of the cohort required daily help with taking medications has implications for the targeting of educational interventions. For those elderly patients experiencing readmissions for CHF, the risk of dying in the near future and the palliative nature of treatment are additional issues which may need to be explored.

This study was constrained, for practical reasons, by reliance on a clinical diagnosis of CHF. In a larger study with more resources, verification of diagnosis and more detailed examination of diagnostic coding practices would be desirable, and would allow identification of subgroups of particular clinical interest.

For a major public health problem, CHF presents practical difficulties for monitoring. Medical records under-identify the problem by the practice of coding CHF-related admissions to underlying conditions. Patients who are admitted to hospital with an acute exacerbation of CHF receive a variety of admission diagnoses. Identifying CHF admissions by diagnostic coding currently underestimates the size of the problem, and routine monitoring will require a less labour-intensive approach.

In this study, poor health represented a further barrier to both study participation and monitoring outcomes. Most losses to follow-up resulted from death or worsening health. During the study 21 subjects (17.2%) died. This represents a poor survival rate, worse than for many malignancies of adulthood. Cause of death was not ascertained, but data from the SOLVD (Studies of Left Ventricular Dysfunction) Registry cohort12 shows that most deaths were due to progressive CHF. In addition, sudden death occurs in CHF patients at five times the rate in the general population of the same age.17

In summary, chronic CHF represents a significant burden to patients (through morbidity and mortality), their carers (through provision of daily care), and the hospital system (through multiple admissions for acute decompensation).  

Acknowledgements

The authors wish to acknowledge Sr Moira Hewitt and Sr Jeanette Bunn for their contribution to the Westmead CHF Outcomes study, and study co-investigator Dr Julia Lowe from the Newcastle Heart Failure Group. This study was funded by the NSW Health Department's Health Outcomes Program.  

References
  1. Mortality from Congestive Heart Failure - United States, 1980-1990. Morb Mortal Wkly Rep February 11, 1994;43(5):77-81.
  2. Yamani M,Massie BM. Congestive Heart Failure: Insights from epidemiology, implications for treatment. Mayo Clin Proc 1993;68:1214-1218.
  3. McMurray,McDonagh CE,Morrison CE,Dargie HJ. Trends in hospitalisation for heart failure in Scotland 1980-1990. Eur Heart J 1993; 14:1158-1162.
  4. Editorial. Lancet 1992;330:278-279.
  5. Gooding J, Jette AM. Hospital readmissions among the elderly. J Am Geriatr Soc 1985;33:595-601.
  6. National Centre for Health Statistics, Graves EJ, National Hospital Discharge Survey: Annual Summary, 1990. Vital Health Stat [13] 1992;112:28.
  7. Munoz E,Chalfin D,Birnbaum E,Mulloy K,JohnsonH,Wise L. Hospital costs, resource characteristics, and the dynamics of death for patients with a primary diagnosis of congestive heart failure. n Y State J Med 1989;89:60-63.
  8. The economic impact of heart failure on the UK National Health Service. Eur Heart J 1993;14(suppl):133.
  9. Braunwald E. ACE inhibitors - a cornerstone of treatment of heart failure. n Engl J Med 1991;325:351-353.
  10. SOLVD Investigators. Effect of Enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. n Engl J Med 1991;325:293-302.
  11. Parameshwar J,Poole-Wilson PA,Sutton GC. Heart failure in a district hospital. J. Royal Coll Physicians Lond 1992;26:139-142.
  12. Bourassa MG,Gurne O,Bangdiwala SI,Ghali JK, Young JB, Rousseau M,Johnstone DE, Yusuf S for the SOLVD investigators. Natural history and patterns of current practice in heart failure. J Am Coll Cardiol 1993;22[Suppl A]:14A-19A.
  13. Folstein MF,Folstein SE,McHugh PR. 'Mini-Mental State':a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189-198.
  14. Ware JE,Snow KK,Kosinski M,Gandek B. SF-36 Health Survey: manual and interpretation guide. Boston, Massachusetts 1993: the Health Institute, new England Medical Centre.
  15. SAS Institute Inc. SAS for the Microsoft Windows Environment, Version 6. Cary, NC: SAS Institute Inc , 1993.
  17. Stevenson CE 1996. SF-36:Interim norms for Australian data. Canberra:Australian Institute of Health and Welfare.
  18. Kannel WB,Ho K,Thom T. Changing epidemiologic features of cardiac failure. Br Heart J 1994;72(suppl):S3-S9.

(Received 8 Oct 1996, accepted 10 Apr 1997)

 

Authors' details

Westmead Hospital, Westmead, NSW 2145
Fiona M Blyth, FAFPHM, Registrar, Department of Public Health and Community Medicine.
Ross Lazarus, FAFPHM, Senior Lecturer in Epidemiology, Department of Public Health and Community Medicine.
David Ross, FRACP, Head, Department of Cardiology.
Michael Price, FRACP, Head, Geriatric Medicine Unit.
Gary Cheuk, FRACP, Registrar, Geriatric Medicine Unit.
Stephen R Leeder, FRACP, PhD, FAFPHM, Professor of Public Health and Community Medicine.


MJA reviewers' comments - Register to be notified of new articles by e-mail - To Papers for review list - To top of article - ©MJA1997

<URL: http://www.mja.com.au/> © 1997 Medical Journal of Australia.