Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
heart failure and south asians.
- dr.manan vasenwala (8 September 2003)
-
Caution please
- Kiran C Patel (9 September 2003)
-
Deprivation gradient in mortality should not be dismissed as artefactual
- Georgios Lyratzopoulos, Richard F Heller (25 September 2003)
-
Asians with heart failure die younger
- Michael D Sosin, Jeetesh V Patel, Gurbir S Bhatia, Elizabeth A Hughes, Derek L Connolly, Russell C Davis (3 October 2003)
-
Counterintuitive findings on heart failure in South Asians: consider artefacts
- Raj S. Bhopal, Colin Fishbacher (4 November 2003)
|
|
|||
|
dr.manan vasenwala, consultant-cardiologist (non-invasive) k.k.heart center, aligarh-202002.india
Send response to journal:
|
this study brings to prominence again what has been observed for a long time that ihd occurs at least a decade earlier in asians than in whites. it is not a reflection of the age of the local population as suggested by the authors. although the major risk factors are common to both the groups, novel risk factors like lipo-protein 'a ' or homocystein levels may be responsible. the habit of tobocco chewing may play a part. genetic studies may be needed to unravel the mystery like susceptibility and therapeutic response genes in coronary disease. Various common genotypes and haplotypes have been identified that are relevant to hyperlipidemia, vascular tone and blood pressure, metabolic syndrome and diabetes, homocysteine levels, smoking behavior, endothelial and arterial wall integrity, plaque development and stability and inflammatory response. but as of now, no such studies are available in south asians. Competing interests: None declared |
|||
|
|
|||
|
Kiran C Patel, SpR inCardiology Queen Elizabeth Hospital, Birmingham B15 2TH
Send response to journal:
|
Dear Editor, Whilst the results of Blackledge et al provide an interesting insight, let us stress that the patients in this study were hospital inpatients. One is aware that many deaths from heart failure occur suddenly and in the community and hence those that make it to hospital, as in this study, are a self-selected group of 'survivors'. It may be possible that mortality rates from heart failure in South Asians are higher in the community than in European Whites and this study in no way addresses this possibility. Of great concern is the finding that conclusions stated in some of the recent media following this study, that survival rates from heart failure in South Asians is better than the general population, are incorrect and breed false optimism. Though a study of heart failure in ethnic subgroups is commendable, a large scale study, extending into the community, would provide us with a more accurate picture of heart failure in South Asians. One would predict that, similar to coronary heart disease, the disease occurs in younger age groups and has an overall greater incidence, prevalence, morbidity and mortality in South Asians when compared to European whites. Competing interests: None declared |
|||
|
|
|||
|
Georgios Lyratzopoulos, Lecturer in Public Health Evidence for Population Health Unit, 2nd Floor Stopford Building, Oxford Road, Manchester, M13 9PT, Richard F Heller
Send response to journal:
|
We are grateful to Blackledge, Newton and Squire for their important study of prognostic variability in heart failure patients of different ethnic origin (1). An interesting finding was that, counter-intuitively, more socially deprived heart failure patients have a better all-cause mortality outcome than less deprived patients. This contradicts the findings of a large previous UK study showing a clear socioeconomic gradient in mortality risk in favour of the least deprived patients (2). The authors suggest that their finding may be artefactual, due to the deprivation index used (Index of Multiple Deprivation 2000). This is indeed possible. However inspection of the data shows a systematic socioeconomic mortality risk gradient in favour of the more deprived. It is notable that although the results as presented are only statistically significant for the most deprived group, performing a test for trend by using the deprivation score as a continuous variable is likely to have produced a statistically significant result and can perhaps have been more informative (3). Misclassification error that could result from the use of any ecological deprivation index would have been expected to influence results towards parity, rather than produce a clear socioeconomic gradient in either direction. As the authors point out, adverse health outcomes are indeed concentrated in elderly population sub-groups of any given geographically defined population. In spite of this fact, area-based UK deprivation indices have been shown consistently to be predictors of poor health outcomes at the individual level (4). Therefore to readily attribute the findings to misclassification error due to the deprivation index used may be wrong. An alternative hypothesis is that the observed prognostic differences in favour of more deprived patients might reflect differences in the underlying causes of heart failure in the different deprivation groups. More deprived groups may contain relatively more people with heart failure due to causes such as valve disease, hypertension, alcoholism and arrhythmias and fewer with coronary artery disease -which has the poorer prognosis (5). The exact causes of the socioeconomic differences in mortality observed in this study merit further investigation and should not be dismissed as artificial in the first instance. G Lyratzopoulos, Lecturer in Public Health, Evidence for Population Health Unit, School of Epidemiology and Health Sciences, University of Manchester. RF Heller, Professor of Public Health, Evidence for Population Health Unit, School of Epidemiology and Health Sciences, University of Manchester. 1. Blackledge HM, Newton J, Squire IB. Prognosis for South Asian and white patients newly admitted to hospital with heart failure in the United Kingdom: historical cohort study. BMJ 2003;327:526-30. 2. MacIntyre K, Capewell S, Stewart S, Chalmers JWT, Boyd J, Finlayson A, Redpath A, Pell JP, McMurray JJV. Evidence of improving prognosis in heart failure. Trends in case fatality in 66,547 patients hopsitalized between 1986 and 1995. Circulation 2000;102:1126-1131. 3. Clark TG, Badburn MJ, Love SB, Altman DG. Survival Analysis Part IV: Further concepts and methods in survival analysis. Br J Cancer 2003;89:781 -786. 4. Carstairs V. Deprivation indices: their interpretation and use in relation to health. J Epidemiol Community Health 1995;49 Suppl 2:S3-8 5. Cowie MR, Wood DA, Coats AJS, Thompson SG, Suresh V, Poole-Wilson PA, Sutton GC. Survival of patients with a new diagnosis of heart failure: a population based study. Heart 2000;83:505-510. Competing interests: None declared |
|||
|
|
|||
|
Michael D Sosin, Research Fellow, Dept of Cardiology Sandwell & West Birmingham NHS Trust, West Bromwich B71 4HJ, Jeetesh V Patel, Gurbir S Bhatia, Elizabeth A Hughes, Derek L Connolly, Russell C Davis
Send response to journal:
|
We wholly agree with Blackledge et al. [1], that there is a need for further research on heart failure amongst the minority ethnic communities. However, as heart failure is a difficult condition to diagnose (and therefore study), facilities to confirm case definition, based on current NICE guidelines would not have been available for all subjects. Given the insidious onset, chronic duration and long-term therapy, morbidity rather than mortality might be more suited in the assessment of disease burden. Furthermore, it is unlikely that hospital admission data is sensitive to the consideration that not all patients present with typical symptoms of heart failure. Many patients may be diagnosed on presentation of other illness such as myocardial infarction, and determining the true onset of heart failure may not be very accurate. Cultural differences in disease perception and understanding may be important factors amongst ethnic minorities, further compounded by language barriers. These factors may confound the survival data of Blackledge et al. We note that the authors quote lower risks of death among South Asian patients of both sexes. However, the unadjusted death rate at all measured time points (30 days, 1 year, and 2 years) were only significantly different in females aged less than 75. Even if survival is greater among South Asian patients, which has not been convincingly demonstrated here, the author’s optimistic conclusion rather misses the point: most South Asian patients presenting with heart failure will be dead before their Caucasian same age counterparts begin to experience symptoms. We have analysed data from death certifications (International Classification of Disease, 9th revision) in the multi ethnic region of Sandwell (West Midlands, UK). A total of 18299 deaths were recorded between 1995-1999. Country of birth was used to determine ethnicity. Age- adjusted deaths due to coronary heart disease (CHD) were proportionately more common among Indian women (27.6% [95%CI: 19.4-35.8]), when compared with the indigenous population (19.2% [18.4-20.1]). Of note, the mean age of death amongst women with a reported co-morbidity of heart failure, who were Indian (72.8 yrs [69.0-76.5]) or Caribbean (69.6 yrs [63.4-75.8]) was significantly lower than that for indigenous females (80.1 yrs [79.5- 80.7]). This was also apparent when Indian and indigenous males were compared: 67.4yrs [64.1-70.7] vs.73.6yrs [72.9-74.2]. Could an earlier onset of heart failure amongst South Asians explain marginally better survival rates? Blackledge’s heart failure survival data should not detract from the fact that South Asians have a disproportionately high burden CHD burden compared to the general UK population. Unfortunately, epidemiological and therapeutic studies in heart failure performed to date have included few patients of South Asian ethnicity. Efforts must be made to ensure that future trials address the population at risk. [1] Blackledge HM, Newton J, Squire IB. Prognosis for South Asian and white patients newly admitted to hospital with heart failure in the United Kingdom: historical cohort study. BMJ 2003;327:526-30 Competing interests: None declared |
|||
|
|
|||
|
Raj S. Bhopal, Professor Public Health Sciences, University of Edinburgh, Teviot Place, Edinburgh, EH8 9AG,, Colin Fishbacher
Send response to journal:
|
Editor, Blackledge et al’s innovative paper on heart failure among UK South Asians is a welcome contribution. (1) As the authors recognise, some of their findings are counter-intuitive. We suggest that data artefacts might partly explain their findings, as presented in the internet version. They report a huge excess of hospital admissions for heart failure in South Asians, and yet a better outcome despite more diabetes and hypertension, and better outcomes among those with greater levels of deprivation. Some of the results might be explained by inconsistent numerators and denominators, one of the most important causes of counterintuitive findings in epidemiology. (2) The analysis of hospital admissions uses cases from 1998-2001 but a denominator population from the 1991 Census. We assume that 2001 census data was not available at the time the analyses were done, but could this mismatch explain some of the excess of admission in South Asians? We wonder whether the ethnic codes used in hospital data were the same as those used in the 1991 census, and whether the populations combined in the composite category “South Asian” were the same in the numerator and denominator. It is surprising that the crude incidence rates were substantially raised in heart failure for any admission but only marginally raised for first admissions. Can the authors justify their statement that the limitations of hospital discharge data apply equally to both ethnic groups? We note that Table 1 shows 85% of South Asian patients lived in the most deprived areas (Q5), compared to 38% of white patients, suggesting substantial geographical clustering. There are typographical errors in table 2 (e.g. in the row relating to atrial fibrillation, and the numbers in the subcategories under acute myocardial infarction). Figure 1 shows an age adjusted ratio for heart failure admission of about 2.8 in men and 4.3 in women, in apparent contradiction to the figures given in the abstract (3.8 and 5.2). We offer three alternative and testable potential explanations for these unusual findings. (a) South Asians’ huge excess of heart failure, aside from the contradictory data in abstract and figures, may be largely an artefact of inappropriate denominators. These excesses are out of proportion to the modest excess in coronary mortality or morbidity in South Asians (3). One possibility is that South Asians live in the inner city close to the local hospitals, while white patients are scattered across the city and are less likely to be admitted to hospital with heart failure, or are more likely to be admitted to hospitals outside Leicestershire. (b) The better outcome in South Asians results from their younger age. As the authors recognise, the younger age reflects their younger age structure in the population i.e. demography, not pathology. The statement that “the younger age reflects the earlier onset of disease” is therefore misleading since the source population is younger. It is not easy to control for age when there are big differences, and there may be residual confounding. The findings might be tested by restricting the outcome analysis to comparable age groups. (c) The better outcome in the most deprived quintile reflects the high proportion of younger South Asians, with incomplete control of ethnic group and age as potential confounding factors. We look forward to the authors’ response. References 1. Blackledge H M, Newton J, and Squire I B. Prognosis for South Asian and white patients newly admitted to hospital with heart failure in the United Kingdom: historical cohort study. BMJ 327:1-6, 2003. 2. Bhopal R. Concepts of Epidemiology. Oxford, OUP, 2002. 3. Bhopal R. What is the risk of coronary heart disease in South Asians? A review of UK research. Journal of Public Health Medicine 2000:22;375- 385 Competing interests: None declared |
|||