 
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
28-January-2005, Vol 118 No 1208
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Heart failure hospitalisations and deaths in New Zealand:
patterns by deprivation and ethnicity
Tania Riddell
Heart failure is a debilitating and costly
disease.1 The aging of populations and enhanced
survival of coronary heart disease patients in developed countries now means
that heart failure is a significant public health concern. Indeed, it is the
only major cardiovascular disease known to have an increasing
prevalence.2
Although heart failure is a common disease, with high
mortality and hospitalisation rates, very little was known of its extent in New
Zealand (NZ) until relatively recently. Hospital admissions for heart failure
increased by more than 50% between 1988 and
1997,3 and were estimated to consume
approximately 1% of the New Zealand health
budget.4 Even less was known of the
socioeconomic and ethnic differences for heart failure outcomes. However, two
studies have demonstrated that deprivation is positively associated with heart
failure hospitalisation rates in New Zealand,5
and that mortality and hospitalisation rates are up to eight times higher
among Maori compared to non-Maori in the age group 45–64
years.6
This study examined the association between socioeconomic
deprivation and heart failure outcomes in Maori and non-Maori New Zealanders for
the period 1988–1998.
MethodsData on deaths
(1988–1997), and public hospital discharges (1988–1998) with heart
failure listed as a primary cause, were obtained from the New Zealand Health
Information Service’s (NZHIS) National Minimum Data Set (NMDS).Details of
the International Classification of Diseases (ICD9) codes used in this study are
presented in Table 1.
Table 1. ICD9 classification of heart failure
deaths/hospitalisations
Source:
International classification of diseases, 9th
revision, 1989
The NMDS does not include a measure of socioeconomic
position. However, domicile of residence is assigned to each record, which
allows an area-based measure of deprivation to be derived. Using the meshblock
code assigned by NZHIS, a NZ Index of Deprivation (NZDep) decile was assigned to
each record. NZDep is a measure of small area deprivation, with decile one being
the least deprived and decile ten the most deprived of small areas in New
Zealand.7
Because Maori are under-represented in the least
deprived deciles, deciles one to four were collapsed to improve statistical
stability. This created seven NZDep strata (1–4, 5, 6, 7, 8, 9, 10) for
comparison within this study. Data were split into the intervals 1988–1992
and 1993–1998 allowing for variation between the NZDep91 and NZDep96
indices.
The 1991 and 1996 census population distributions of
over 45 year old people, by age group, sex, ethnicity, and NZDep91/96 strata
were used as the denominators for the calculation of rates. Direct
standardisation was used to assess the independent associations of ethnicity and
small area deprivation with heart failure outcomes, and to control for
age.8 The total New Zealand population
according to the 1991 and 1996 census populations were used as the external
standards. In a sensitivity analysis, the total Maori population was also used
as an external standard.
Age was measured at death or hospital discharge and
presented as two groups: 45–64 years, and 65 years and older. The
groupings were based on an examination of crude numbers within 10-year age
groups and a judgement of lifestages in relation to heart failure as a chronic
disease.
For the classification of ethnicity, any record that
included Maori in either the NMDS or NHI fields was identified as Maori. All
remaining records, including those with no ethnicity specified, were classified
as non-Maori. The rationale for this priority method of classification has been
described previously.5,6
Poisson regression analysis was also used to determine
the association between ethnicity and heart failure outcome after allowance for
deprivation and age. All analyses were carried out using SAS version
8.0.
ResultsIn total, 8079 heart failure deaths
(1988–1997) and 66,416 heart failure hospitalisations (1988–1998)
for people aged 45 years and older were recorded by the NMDS. Socioeconomic
deprivation was significantly associated with a greater likelihood that the
patient was Maori (p<0.0001).
Heart failure deaths and hospitalisations increased across
deprivation strata. Overall, for an increase of one NZDep decile, there was an
11% increase in heart failure deaths (p<0.01) and hospitalisations
(p<0.0001) (Table 2).
Table 2. Poisson regression analysis of relative risk
for increase of one NZDep decile
Source of
base data: NZHIS.
Within deprivation strata, Maori mortality and
hospitalisation rates were about four to five times higher than those of
non-Maori aged 45–64 years, and twice those of non-Maori aged 65 years and
older. Tables 3 and 4 show the Maori/non-Maori rate ratios and 95% confidence
intervals for males by NZDep96.
Table 3. Maori and non-Maori male heart failure
mortality rates and rate ratios by NZDep96, 1993–1997
Source of
base data: NZHIS.
Note:
rate per 100,000 per year, age-standardised to total New Zealand
population.
Figures 1 and 2 are the graphical representation of these
results. It is clear from the graphs that the relationship is not linear
(probably due to small numbers). However, the age-standardised heart failure
death and hospitalisation rates were higher for Maori compared to
non-Maori.
Figure 1. Heart failure
mortality for males by NZDep96, ethnic
and age group, 1993–1997
 Figure 2. Heart failure
hospitalisation for males by NZDep96,
ethnic and age
 group, 1993–1998
 Table 4. Maori and non-Maori male heart failure
hospitalisation rates
and rate ratios by NZDep96, 1993 - 98
Source of
base data: NZHIS.
Note:
rate per 100,000 per year, age-standardised to total New Zealand
population.
DiscussionThis study has demonstrated that
mortality and morbidity from heart failure in New Zealand has a social gradient.
More deprived groups experience a greater burden of the disease. This is
consistent with the findings of other
studies5,9.
Importantly, what this study adds is that (within any given
socioeconomic strata) Maori heart failure outcomes compare unfavourably with
those of non-Maori. These differences were particularly marked in middle age
(45–64 years). Specifically, Maori aged 45–64 years had four to five
times the heart failure death and hospitalisation rate of non-Maori.
An excess of this magnitude clearly demonstrates that Maori
experience worse heart failure outcomes that are independent of, and in addition
to, those associated with deprivation. Thus, the associations evident in this
study suggest that although deprivation is among the causes of excess heart
failure mortality and morbidity for Maori in New Zealand, it does not in itself
provide a full answer.
These results add to the already robust evidence of
socioeconomic inequalities in cardiovascular disease. Previous studies suggest
that systematic inequalities in the distribution of income, education,
employment, housing, and health services are major determinants of
cardiovascular disease
inequalities.10–13
Several potential explanations for the differences in heart
failure outcomes between Maori and non-Maori should be considered. One
possibility is that the severity and/or underlying pathophysiology of heart
failure differs between Maori and non-Maori. Maori are known to have a higher
prevalence of diseases that predispose to heart failure such as hypertension,
coronary artery disease, diabetes, and valvular
disease.14,15
It is likely these diseases contribute to the early
development of heart failure with greater end organ severity. However, while the
aetiological profile of heart failure in Maori remains unknown this hypothesis
is unproven. Indeed, pathologically it has been found that Maori do not exhibit
a higher prevalence of cardiomyopathy post- myocardial infarction compared to
European New Zealanders.16
A second plausible explanation for this finding may be that
access to care and quality of care differs between Maori and non-Maori. Heart
failure has been identified as a common condition that leads to ‘avoidable
hospitalisations’—hospitalisations that are potentially preventable
by timely and effective health service
interventions.17 This is not a new concept in
health research.12However, there is very little
New Zealand evidence to substantiate or refute this hypothesis.
The interaction of socioeconomic position with ethnicity,
and its association with cardiovascular health, is complex and not adequately
understood. Socioeconomic and ethnic inequalities in health are the consequence
of a lifetime of exposure to disadvantage.18
These inequalities differentially shape population groups’ exposures to
cardiovascular disease risk factors (such as
hypertension,19
smoking,20 physical
activity,21 and
obesity22) and their access to and utilisation
of health services.23
Neighbourhood influences may provide an additional pathway
through which social structures shape cardiovascular disease risk
exposure.24 This is of interest within the New
Zealand context where the impact of living in disenfranchised and impoverished
communities impacts more on Maori than non-Maori New
Zealanders.25
Racism has been described as the ‘missing
variable’ within ethnic health disparities
research.26 Evidence of ethnic inequalities in
health and health care are abundant, but their underlying causes, and especially
the contribution of racism are controversial. Indeed, the health professions,
governed by ethical codes of duty, find the charge of racism hard to
bear.27
Finally, it is plausible that all of these explanations may
contribute to the observed ethnic difference in heart failure outcomes in New
Zealand.
Several problems related to the methods of this type of
study have been described previously.28 It was
a retrospective study that used administrative datasets, and had the inherent
limitations of this design. Thus, it was prone to confounding and its
cross-sectional nature provided no basis for determining causal directionality.
Such studies are often criticised for expecting
administrative data to provide more than they
should.29 Two sources of bias must be
discussed. First, misclassification of ethnicity data leads to the
under-enumeration of Maori within official health datasets. This in turn
produces an under-estimation of Maori rates. Therefore, the rates calculated in
this study are likely to be an underestimate of the true mortality and morbidity
associated with Maori heart failure.At the same time, census data are known to
undercount Maori, although modest undercounts of Maori in the census would be
unlikely to have meaningfully changed the overall patterns observed in this
study. Second, misclassification of ICD9 codes had the potential to contribute
to bias within this study. If access to echocardiography is different for Maori
compared with non-Maori, this could bias the diagnosis of heart failure and
hence measurement of rates. Again the likely effect of this potential bias, if
it existed, would have been to underestimate the true mortality and morbidity
associated with heart failure for Maori in this study.
Some other weaknesses are acknowledged. NZDep does not
adequately capture lifecourse experience. Area-based measures may be less
adequate than individual measures to control for socioeconomic confounding in
analyses of ethnic differences in health.30
Lastly, this study could not investigate clinical factors such as disease
severity nor undertake an audit of medical records to confirm the accuracy of
diagnosis.
In conclusion, this study demonstrates that socioeconomic
deprivation is associated with an increased chance of death and hospitalisation
from heart failure in New Zealand. Importantly, within socioeconomic strata,
Maori disparities in heart failure outcomes persist. Clarification of the
explanations for these results await more detailed studies—a call to
action in the interim is required, however.
Specifically, a broad approach that includes structural
interventions (such as social policy to address ethnic power relations and
racism in New Zealand), social interventions (such as policy to increase Maori
access to affordable and high quality housing) and health sector interventions
(such as measures to increase Maori access to culturally competent health
services and boost the Maori cardiovascular workforce) is required.
Author
information: Tania Riddell, Public Health Specialist, Tomaiora Maori
Health Research Centre, Auckland Medical School, University of Auckland,
Auckland
Acknowledgements: I
thank Papaarangi Reid and Peter Crampton for their supervision and advice. I am
also grateful to my colleagues at Tomaiora Maori Health Research Unit (Auckland
School of Medicine) and Te Ropu Rangahau Hauora a Eru Pomare (Wellington School
of Medicine). Further acknowledgement goes to the Australasian Faculty of Public
Health Medicine for supporting me with a John McLeod scholarship.
Correspondence:
Dr Tania Riddell, Public Health Specialist, Tomaiora Maori Health Research
Centre, Auckland Medical School, University of Auckland, Private Bag 92019,
Auckland. Fax: 09 376 3238; email: t.riddell@auckland.ac.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|