Q J Med 2003; 96: 919-923
© Association of Physicians 2003; all rights reserved.
Ethnic trends in lipid tests in general practice
From the 1Department of Clinical Biochemistry, Manchester Royal Infirmary, 2Barlow Medical Centre, Manchester, 3Department of Epidemiology, University of Manchester, and 4Department of Clinical Biochemistry, Manchester Royal Infirmary, Manchester, UK
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: South Asian migrants have a higher cardiovascular mortality than the indigenous population. Contributory factors may include a lower rate of cardiovascular risk assessment uptake and compliance.
Aim: To compare rates of lipid testing, follow-up and patterns of dyslipidaemia in South Asian and non-South-Asian populations in Manchester.
Design: Retrospective cross-sectional survey.
Methods: Lipid requests from 14 general practices were analysed, using name recognition software to assign ethnicity.
Results: Compared with non-South-Asians, the age-standardized rate of lipid testing in South Asians was similar in men at 12.1% (95%CI 11.6–12.6) vs. 11.2% (9.5–13.0), but higher in women at 18.6% (15.9–21.2) vs. 13.2% (12.6–13.7). Trends of cholesterol with repeat testing were similar in the two populations. However, South Asian women had lower mean levels of total cholesterol (5.50 vs. 5.68 mmol/l, p = 0.021), lower levels of HDL (1.20 vs. 1.46 mmol/l, p < 0.001), an excess of hypertriglyceridaemia (1.62 vs. 1.45 mmol/l, p = 0.035) and a greater proportion with cholesterol > 5.2 mmol/l combined with low HDL (43.1% vs. 20.2%, p = 0.002). South Asian men had lower levels of total cholesterol (5.17 vs. 5.37 mmol/l, p = 0.048) and lower levels of HDL (1.07 vs. 1.64 mmol/l, p < 0.001). There was no difference in the proportion of South Asians men and women with cholesterol < 5.2 mmol/l combined with low HDL.
Discussion: The rate of lipid testing and change in cholesterol levels with repeated testing did not differ between South Asian and non-South-Asian groups. The pattern of dyslipidaemia seen in this South Asian population, especially women, was different from that of the non-South-Asian population, with possible implications for cardiovascular risk assessment.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
There is an excess of cardiovascular morbidity and mortality in people of South Asian ethnic origin in the UK,1–11 which persists despite current strategies of risk factor management. This may reflect an excess of known major risk factors or other factors such as homocysteine,12 dietary fat13 or lipoprotein(a)14–17 acting disproportionately on this ethnic group. However, it may also reflect inadequate cardiovascular risk factor modification. In this study, we compared routine lipid testing data deriving from general practitioner requests stored on the Clinical Biochemistry laboratory computer system to study differences in lipid testing between South Asians and non-South-Asians. We assessed the proportion of the two groups being tested, the rate of re-testing, change in mean lipid levels over time and the prevalence of patterns of lipid results with particular significance for coronary heart disease (CHD) risk assessment.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study was restricted to 14 general practices that use the Biochemistry Department at Manchester Royal Infirmary. Manchester Health Authority provided us with the names, gender and ages of all patients registered with the 14 general practices. Lipid data on all patients aged > 30 years were extracted from the laboratory computer system, together with patient name, gender and age. Ethnicity was assigned on the basis of name recognition using the Nam Pehchan computer programme18 developed by Bradford Health Authority, which has been shown to categorize correctly 90% of patients into South Asian or non-South-Asian when applied to communities living in the North-West of England. The study period was 2 years from 1 July 1999 to 30 June 2001. Patients were excluded from the study if they had a cholesterol measurement recorded within the 2 years preceding the first measurement in the study period. After assigning ethnicity, patient identification was coded in order to anonymize the database permanently. The proportions of South Asians and non-South Asians having cholesterol tests were compared after standardizing for age. Combined trends in cholesterol for men and women in the two ethnic groups were assessed in those having repeated measurements. We assessed the proportion of the South Asian population with low HDL and high TG (insulin-resistant phenotype) according to level of total cholesterol above or below 5.2 mmol/l. The insulin resistance phenotype was defined by us as an HDL
1.0 mmol/l for men and 1.3 mmol/l for women, associated with triglycerides 
1.7 mmol/l.
Lipid measurements
All measurements were made routinely using the standard laboratory protocols and procedures. Serum cholesterol was measured using the Roche CHOD-PAP method, HDL cholesterol by the Roche HDL-C Plus homogenous method using magnesium sulphate and dextran sulphate to exclude non-HDL cholesterol and triglycerides by the Roche GPO-PAP enzymic method on a Roche modular instrument (Roche Diagnostics). The laboratory participated in the National External Quality Assurance Scheme (Wolfson EQA laboratory) and the Welsh National External Quality Assurance Scheme (UHW). General advice is given to general practitioners that diagnostic lipid profiles should be measured in the fasting state, but no specific measures were taken to validate this. This study was conducted according to the criteria for medical audit approved by the Central Manchester Research Ethics Committee. No file of identifiable patient data was created for the purpose of this investigation.
Statistical methods
Age-standardized means and prevalence of levels of biochemical measurements were calculated for South Asian and non-South-Asian men and women by the direct method, using the age distribution of those who had their first cholesterol measurement in the study period. Statistical differences of the means of lipid levels between the two ethnic groups were tested using age-adjusted linear regression, with the biochemical measurement as the outcome variable and ethnic group as the dependent variable. Statistical differences of percentages between groups were tested using age-adjusted logistic regression, with the lipid category as the outcome variable and the ethnic group as the dependent variable. Trends in cholesterol with repeated measurement were estimated using generalized least-squares regression. All analyses used the Stata statistical package (Timberlake Consultants).
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Approximately 35 000 people aged 30 years or older were registered with the 14 general practices. The Nam Pehchan program identified 3700 (10.3%) of these as being of South Asian origin. First cholesterol measurements were recorded in the study period for 3573 people from this population, and Nam Pehchan identified 348 (9.7%) as South Asians. The rate of cholesterol measurement was similar for South Asian and non-South-Asian men: 12.1% (95%CI 11.6–12.6) vs.11.2% (9.5–13.0). But South Asian women had a higher age-standardized rate of testing: 18.6% (95%CI 15.9–21.2) vs. 13.2% (12.6–13.7). The sex-specific age-standardized means for lipid levels and prevalence rates of lipid parameters in South Asians and non-South-Asians are shown in Table 1.
|
Although South Asian men and women had lower total cholesterol, a higher proportion of them also had lower HDL. The proportions of South Asian and non-South-Asian men and women having an insulin resistance lipid phenotype, stratified according to level of total cholesterol, are shown in Table 2. This lipid pattern was seen in a higher proportion of South Asian women (33.4%) compared with non-South-Asian women (19.0%) (p = 0.008), and the difference was most marked in women with total cholesterol levels > 5.2 mmol/l.
|
The mean cholesterol levels for all subjects in the two ethnic groups according to the number of times they had their cholesterol measured within the study period is shown in Table 3. About a fifth of patients in each group had their cholesterol measured a second time within the two-year study period. There was no difference in the proportion being retested in the two ethnic groups (p = 0.867). There was an average reduction in total cholesterol per repeat measure of 0.25 mmol/l (95%CI 0.20–0.30) among non-South-Asian men and women combined who had their cholesterol measured more than once during the study. The reduction was 0.13 mmol/l (95%CI 0.01–0.26) among South Asian men and women combined.
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The primary purpose of this study was to use routinely collected laboratory data to assess rates of lipid testing, frequency of repeat testing, trends in cholesterol over time and lipid profile patterns in the South Asian population compared with the rest of the predominantly Caucasian population. The laboratory database does not include ethnicity, distinguish between primary and secondary prevention, or record whether the patient is on cholesterol-lowering treatment. Although self-assessment is the preferred way of assigning ethnicity, name recognition software has been shown to have reasonable accuracy, and was applied to those tested as well as the comparison general practice population. The overall Manchester population is 89% Caucasian, 6.5% Pakistani and 4.5% Black Caribbean or Black African, with small numbers of other ethnic groups. Although it would be of interest to study the Black population as a separate group, we had no means of assigning ethnicity on the laboratory database. However, Black Africans and Black Caribbeans are a minor component of the non-Asian population, and therefore will have a small effect. The analysis was weighted in favour of newly initiated cholesterol testing activity by restricting it to patients for whom there was no cholesterol measurement on the database for 2 years preceding the study period. Although not likely to be completely accurate, the same selection was made for both population groups.
We found no difference between the age-standardized rates of testing for men. Similar rates of testing were found for men in both groups and in non-South-Asian women. However, in South Asian women, a higher rate of testing was found. This increase is unexplained. Overall, these results indicate that there is no significant under testing of this South Asian population.
Although the mean trend for cholesterol reduction in South Asians was lower than non-South-Asians, the confidence interval was wide for the Asian group, because they only represent about 10% of the population and the sample size was relatively small. There was no difference in the rate of re-testing between the groups. This could be a useful measure if applied to large numbers.
The lipid pattern observed in South Asians in this Manchester community is similar to that seen in other communities in UK cities,19–23 with lower average total cholesterol, lower average HDL and a tendency towards hypertriglyceridaemia. The latter two features are characteristic of the insulin resistance lipid phenotype, which is more prevalent in South Asians.24 Concern exists as to whether the Framingham algorithm for cardiovascular risk estimation is generalizable to non-Caucasian populations with different prevalences of risk factors and underlying rates of CHD. Overestimation of risk was found when applying the algorithm to ethnic minority groups in the USA.25 No studies have been done in South Asians, in whom an underestimation of risk compared with Caucasians would be expected, due to a higher prevalence of heart disease. Use of the cholesterol/HDL ratio in the algorithm may underestimate the contribution of low HDL to risk when cholesterol is low. Recalibration of the Framingham algorithm for different mean risk factor prevalences and use of HDL and cholesterol as separate variables has been suggested.25 Application of the Framingham algorithm to South Asians living in London has been shown to identify a higher proportion of individuals at high CHD risk, with a disproportionately high rate of CVD risk compared with the Caucasian population.23 Thus recalibration and use of HDL as a separate variable may identify an even higher proportion at risk. We found a significantly greater proportion of South Asian women with cholesterol > 5.2 mmol/l had an insulin-resistant lipid phenotype. There was no significant difference between the proportions of South Asian men and women with cholesterol < 5.2 mmol/l compared with non-South-Asians. This study suggests that our local South Asian population is not disproportionately liable to underestimation of cardiovascular risk due to the combination of low cholesterol and HDL compared with the local non-South-Asian population. However, the higher prevalence of low HDL in South Asians implies that recalibration may improve the accuracy of risk estimation, particularly in women.
Our data on predominantly Pakistani Manchester South Asians cannot be generalized to other populations. Patterns of lipid abnormality in South Asians vary quite markedly within South Asian sub-populations,26 and may well vary with geographical location within the UK. The pattern is likely to change with acculturation, as has been shown for Punjabis living in West London27 and in second-generation South Asians living in Newcastle.28 Therefore, local initiatives for monitoring ethnic minorities are needed. Laboratory data bases, which routinely store lipid data, would be more powerful tools to monitor interventions and trends, if data on ethnicity and drug treatment were collected with lipid requests. This pilot study, using proxy data for ethnicity and treatment status, has indicated the potential usefulness of this approach. One important advantage of using routinely collected data is that repeated analysis can be made on large volumes of data with little additional effort or cost. UK guidelines for taking ethnicity into account for CHD risk assessment remain vague. There remains a danger that an inappropriately low level of risk factor modification will persist in the South Asian community.
|
Acknowledgments |
|---|
We are grateful to Central Manchester Primary Care Trust for its support of this work by a Health Gain Fund grant. MWF, SK and CJS conceived the study. PM advised on the study design and performed the statistical analyses. All the authors contributed to the manuscript.
|
Footnotes |
|---|
Address correspondence to Dr M.W. France, Department of Clinical Biochemistry, Manchester Royal Infirmary, CMMCUH NHS Trust, Oxford Road, Manchester M13 9WL. e-mail: mfrance{at}labmed.cmht.nwest.nhs.uk
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Balarajan R. Ethnic differences in mortality from ischaemic heart disease and cerebrovascular disease in England and Wales. Br Med J 1991; 302:560–4.
2. Enas EA, Yusuf S, Mehta JL. Prevalence of coronary artery disease in Asian Indians. Am J Cardiol 1992; 70:945–9.[CrossRef][Web of Science][Medline]
3. Pais P, Pogue J, Gerstein H, Zachariah E, Savitha D, Jayprakash S et al. Risk factors for acute myocardial infarction in Indians: a case-control study. Lancet 1996; 348:358–63.[CrossRef][Web of Science][Medline]
4. Cappuccio FP. Ethnicity and cardiovascular risk: variations in people of African ancestry and South Asian origin. J Hum Hypertens 1997; 11:571–6.[CrossRef][Web of Science][Medline]
5. Cappuccio FP, Cook DG, Atkinson RW, Strazzullo P. Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in south London. Heart 1997; 78:555–63.
6. Bhopal R, Unwin N, White M, Yallop J, Walker L, Alberti KG, et al. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study. Br Med J 1999; 319:215–20.
7. Anand SS, Yusuf S, Vuksan V, Devanesen S, Teo KK, Montague PA, et al. Differences in risk factors, atherosclerosis, and cardiovascular disease between ethnic groups in Canada: the Study of Health Assessment and Risk in Ethnic groups (SHARE). Lancet 2000; 356:279–84.[CrossRef][Web of Science][Medline]
8. Game FL, Jones AF. Ethnicity and risk factors for coronary heart disease in diabetes mellitus. Diabetes Obes Metab 2000; 2:91–7.[CrossRef][Web of Science][Medline]
9. Levy D, Kannel WB. Searching for answers to ethnic disparities in cardiovascular risk. Lancet 2000; 356:266–7.[CrossRef][Web of Science][Medline]
10. Bhopal R, Hayes L, White M, Unwin N, Harland J, Ayis S, et al. Ethnic and socio-economic inequalities in coronary heart disease, diabetes and risk factors in Europeans and South Asians. J Public Health Med 2002; 24:95–105.
11. Stewart SH, Silverstein MD. Racial and ethnic disparity in blood pressure and cholesterol measurement. J Gen Intern Med 2002; 17:405–11.[CrossRef][Web of Science][Medline]
12. Seshadri N, Robinson K. Homocysteine and coronary risk. Curr Cardiol Rep 1999; 1:91–8.[Medline]
13. Harper CR, Jacobson TA. The fats of life: the role of omega-3 fatty acids in the prevention of coronary heart disease. Arch Intern Med 2001; 161:2185–92.
14. Bhatnagar D, Anand IS, Durrington PN, Patel DJ, Wander GS, Mackness MI, et al. Coronary risk factors in people from the Indian subcontinent living in west London and their siblings in India. Lancet 1995; 345:405–9.[CrossRef][Web of Science][Medline]
15. Enas EA. Hypertriglyceridemia and elevated lipoprotein(a) are major risk factors for coronary events in middle-aged men. Am J Cardiol 1996; 78:859–60.[Medline]
16. Misra A. Atherosclerosis in Indians and lipoprotein (a). J Assoc Physicians India 1999; 47:313–17.[Medline]
17. Enas EA. Lipoprotein(a) is an important genetic risk for coronary artery disease in Asian Indians. Am J Cardiol 2001; 88:201–2.[CrossRef][Web of Science][Medline]
18. Cummins C, Winter H, Cheng KK, Maric R, Silcocks P, Varghese C. An assessment of the Nam Pehchan computer program for the identification of names of south Asian ethnic origin. J Public Health Med 1999; 21:401–6.
19. Anand SS, Yusuf S, Vuksan V, Devanesen S, Teo KK, Montague PA, et al. Differences in risk factors, atherosclerosis, and cardiovascular disease between ethnic groups in Canada: the Study of Health Assessment and Risk in Ethnic groups (SHARE). Lancet 2000; 356:279–84.[CrossRef][Web of Science][Medline]
20. Bhopal R, Hayes L, White M, Unwin N, Harland J, Ayis S, et al. Ethnic and socio-economic inequalities in coronary heart disease, diabetes and risk factors in Europeans and South Asians. J Public Health Med 2002; 24:95–105.
21. Bhopal R, Unwin N, White M, Yallop J, Walker L, Alberti KG, et al. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study. Br Med J 1999; 319:215–20.
22. Cappuccio FP, Cook DG, Atkinson RW, Strazzullo P. Prevalence, detection, and management of cardiovascular risk factors in different ethnic groups in south London. Heart 1997; 78:555–63.
23. Cappuccio FP, Oakeshott P, Strazzullo P, Kerry SM. Application of Framingham risk estimates to ethnic minorities in United Kingdom and implications for primary prevention of heart disease in general practice: cross sectional population based study. Br Med J 2002; 325:1271–6.
24. Hansen BC. The metabolic syndrome X. Ann N Y Acad Sci 1999; 892:1–24.[CrossRef][Web of Science][Medline]
25. D'Agostino, Grundy S, Sullivan LM, Wilson P. Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation. JAMA 2001; 286:180–7.
26. Bhopal R, Unwin N, White M, Yallop J, Walker L, Alberti G, et al. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi and European origin populations: cross sectional study. Br Med J (Clin Res Ed) 1999; 319:215–20.
27. Bhatnagar D, Anand IS, Durrington PN, Patel DJ, Wander GS, Mackness MI, et al. Coronary risk factors in people from the Indian subcontinent living in west London and their siblings in India. Lancet 1995; 345:405–9.[CrossRef][Web of Science][Medline]
28. Whincup PH, Gilg JA, Papacosta O, Seymour C, Miller GJ, Alberti KG, et al. Early evidence of ethnic differences in cardiovascular risk: cross sectional comparison of British South Asian and white children. Br Med J 2002; 324:635.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Basu, H. Tang, C. E. Lewis, K. North, J. D. Curb, T. Quertermous, T. H. Mosley, E. Boerwinkle, X. Zhu, and N. J. Risch Admixture mapping of quantitative trait loci for blood lipids in African-Americans Hum. Mol. Genet., June 1, 2009; 18(11): 2091 - 2098. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||