Secular trends in the cardiovascular risk profile and mortality ... - Nature

Journal of Human Hypertension (2008) 22, 18–23 & 2008 Nature Publishing Group All rights reserved 0950-9240/08 $30.00 www.nature.com/jhh

ORIGINAL ARTICLE

Secular trends in the cardiovascular risk profile and mortality of stroke admissions in an inner city, multiethnic population in the United Kingdom (1997–2005) A Gunarathne1, JV Patel1, R Potluri1, PS Gill2, EA Hughes1 and GYH Lip1 1

University Department of Medicine, City Hospital, Birmingham, UK and 2Department of Primary Care and General Practice, University of Birmingham Edgbaston, Birmingham, UK

Our objective was to study ethnic differences in the cardiovascular risk profile and mortality of stroke admissions to an inner city teaching hospital serving a multiethnic population in Birmingham, UK, over a 9-year period (1997–2005). Hospital case notes and registry data of 3083 patients admitted with a first onset stroke were reviewed. Secular trends in the prevalence of risk factors (hypertension, diabetes, hyperlipidaemia, atrial fibrillation and myocardial infarction), hospital admission rates and 30-day mortality among Afro-Caribbean, European Caucasian and South Asian ethnic groups were analysed. Between 1997 and 2005, there were 3083 first onset strokes, of whom 47.6% (1595) were men, 9.3% Afro-Caribbean, 57.8% European Caucasian and 15.1% South Asian. There was a significant trend towards a reduction in non-haemorrhagic stroke admissions over the study period (Po0.001), with no ethnic

variation (P ¼ 0.07). Increases in hypertension and hyperlipidaemia were observed (Po0.001), whereas myocardial infarction showed a decline (Po0.001). Compared to other ethnic groups, South Asian patients were younger on admission (Po0.001), had more hyperlipidaemia (Po0.05) and poorer survival at 30 days (P ¼ 002). We conclude that cardiovascular risk profiles among patients admitted with non-haemorrhagic stroke have changed over the last decade. In particular, hyperlipidaemia has increased, especially among South Asians. The reduced decline in stroke admissions and 30-day survival of stroke in South Asians in recent years warrants further investigation and highlights the importance of a targeted health-care approach in the migrant ethnic minorities. Journal of Human Hypertension (2008) 22, 18–23; doi:10.1038/sj.jhh.1002265; published online 2 August 2007

Keywords: hyperlipidaemia; ethnicity; South Asian; Afro-Caribbean; diabetes

Introduction Stroke is a leading cause of premature and preventable death in Britain, and together with its wellknown morbid affliction in survivors, it is a key focus for national health care and research.1,2 The burden of ischaemic stroke is more prominent among the country’s migrant communities,3 and this relates to the impact of its determinants among these ethnic minorities, most notably, hypertension and diabetes mellitus, which are frequently manifest among South Asian and Afro-Caribbean ethnic groups.4–9 Previous epidemiological analyses in Correspondence: Professor GYH Lip, University Department of Medicine, City Hospital, Birmingham B18 7QH, UK. E-mail: [email protected] This paper was dealt with by a guest editor, Associate Professor Bernard Cheung from the University of Hong Kong, and editorial decisions were independent of the Birmingham-based editor. Received 5 May 2007; revised 14 June 2007; accepted 20 June 2007; published online 2 August 2007

the United States have been a valuable insight into the relationship between ethnic variations in ischaemic stroke and the disproportionate distribution of cardiovascular (CVD) risk factors in these groups.10,11 Migrant communities living in the United Kingdom have experienced large lifestyle changes, the effects of which are likely to explain the insidious CVD risk profiles manifest in these ethnic groups.12–14 Historically, some studies among migrant groups in Britain also suggested that some CVD risk factors were absent in these groups, specifically hyperlipidaemia.15 However, there is limited information on how trends in CVD risk factors and stroke in South Asians and Afro-Caribbeans have changed in recent years, particularly so in the current decade.3,16 The objective of the present study was to investigate ethnic differences in the CVD risk profile and mortality of stroke admissions to an inner city teaching hospital serving a multiethnic population in Birmingham, UK. We tested our null hypothesis

Ethnicity, cardiovascular risk and stroke A Gunarathne et al 19

that there were no significant interethnic differences in the CVD risk profile and 30-day mortality of stroke admissions, nor any secular trends that were evident over an 9-year period (1997–2005). We collated data from national mortality registries and hospital records relating to the Sandwell and West Birmingham area, home to a broad multiethnic population.

Methods Using information of our local West Midlands Regional Health Authority computerized Hospital Activity Analysis register, a database of all patients with a first-in-a-lifetime stroke to Sandwell and West Birmingham Hospitals (SWBH) Trust during the period 1997–2005 (International Classification of Disease (ICD) 10th revision, codes 430–438) was developed. This coding system has been proved free of errors for definitive diagnoses,17 although errors are found in the coding of non-specific diagnoses such as ‘viral infections’ or ‘gastroenteritis’. To minimize any inaccuracies in coding, we also cross-referenced with our computerized stroke unit database. Moreover, information from hospital case notes was used to add more detailed information on demography and stroke risk factors (hypertension, diabetes mellitus, atrial fibrillation, myocardial infarction, transient ischaemic attack (TIA), hyperlipidaemia and peripheral vascular disease). Thirty-day case fatality data, obtained from the National Health Tracing Services, were added to this database.18 Classification of stroke type was based on the radiological evidence available for ICD coding at the time of patient discharge or death. Patients with radiological evidence of haemorrhagic strokes and unspecified aetiology were excluded from the study cohort.

Diagnostic criteria and definitions used for coding

Diagnosis of stroke had been defined on clinical symptoms supported by a cerebral imaging (usually a computed tomography (CT) scan) in accordance with the WHO criteria for diagnosis of acute stroke.19 According to our hospital practice, which is based on standard national guidelines,20 each patient admitted with symptoms and signs of stroke will be examined by a senior physician before routine blood tests and special radiological investigations (CT/magnetic resonance imaging (MRI)) for the diagnosis of acute stroke and its pathological type within 48 h. Each patient will undergo at least one electrocardiogram (ECG) recording and blood pressure measurements twice daily. Coding of hypertension was based on the defined blood pressure 4140/90 mm Hg on at least two separate readings. Diabetes diagnosis was based on the relevant clinical history and biochemical evidence of at least two measurements of fasting blood sugar

readings 47.8 mmol/l. Patients with a total cholesterol level 45.2 mmol/l were considered as hyperlipidaemic. Diagnosis of atrial fibrillation depends on the clinical evidence supported by a 12-lead ECG. History of myocardial infarction was based on ECG evidence supported by elevated cardiac troponin levels. Echocardiographic evidence for diagnosis of valve lesions was mandatory to be given a diagnosis of valvular heart disease. Diagnosis of peripheral vascular disease was based on the patient’s clinical symptoms and signs, supported by any radiological evidence of vascular abnormalities. Data analysis

Data were analysed using SPSS version v14 (SPSS Inc., Chicago, IL, USA). The hospital admission rate within a time period was calculated using the total number of hospital admissions for that period, and the stroke admissions for that time as the numerator. Student’s t-test and one-way analysis of variance (ANOVA) tests were used to determine differences between groups with continuous variables and w2 test was used to compare the categorical variables (for example differences in the prevalence of risk factors between secular time points). The prevalence of risk factors was age- and genderadjusted using the age distribution of total hospital admissions (1997–2005). Kaplan–Meier plots were used for univariate survival analysis, and Cox regression models were developed for multivariate analysis.

Results Between 1997 and 2005, there were 3083 hospital admissions, which were first-in-lifetime-strokes, where complete clinical data and mortality data were available. Of these patients, 2405 (78.0%) presented with a non-haemorrhagic stroke, 159 (5.2%) with a haemorrhagic stroke and 519 (16.8%) were unspecified. Of all strokes, 46.6% were men and were significantly younger than women patients (mean (s.d.) 72.1 (10.6) years vs women, 77.1 (10.9) years; Po0.001). South Asians were younger at admission compared to European Caucasians and Afro-Caribbeans (Po0.001; Table 1). Risk factors

Among those patients admitted with non-haemorrhagic stroke, hypertension was the most commonly encountered cardiovascular comorbidity (62.6%), followed by diabetes mellitus (35.0%) and atrial fibrillation (27.4%). Afro-Caribbeans and South Asians had a higher prevalence of diabetes and hypertension compared to European Caucasians (Po0.001), whereas atrial fibrillation and heart failure were more common in the European Caucasian population (Po0.001). South Asians also had a higher prevalence of myocardial infarction and Journal of Human Hypertension


Table 1 Cardiovascular risk profile by ethnic group for patients admitted with non-haemorrhagic strokes between 1997 and 2005 Total sample (n ¼ 2405) Age, years mean (s.d.)

Risk factors, % (95% CI) Hypertension Diabetes Atrial fibrillation Heart failure Hyperlipidaemia Myocardial infarction

T 74.7 (11.0) 62.6 35.0 27.4 11.6 9.2 6.4

M 72.1 (10.6)

F 77.1 (10.9)

South Asian (n ¼ 420) T 68.8 (11.4)

(60.7–64.5) (33.2–36.9) (25.6–29.1) (10.3–12.9) (8.0–10.3) (5.4–7.3)

73.7 50.3 11.8 7.0 10.8 9.1

M 68.3 (11.3) 731

F 69.4 (11.6) 978

(69.5–77.8) (45.6–55.0) (8.7–14.8) (4.6–9.4) (7.9–13.7) (6.4–11.8)

European Caucasian (n ¼ 1709) T 76.7 (10.5) 58.6 29.8 34.8 13.0 9.2 6.2

M 73.4 (10.3) 243

F 79.2 (9.9) 177

(56.3–60.9) (27.6–31.9) (31.9–36.8) (11.4–14.5) (7.8–10.5) (5.1–7.3)

Afro-Caribbean (n ¼ 276) T 71.5 (9.8) 80.0 56.9 9.0 5.7 6.6 5.0

M 71.5 (9.1) 147

P-value*

F 71.4 o0.001** (10.5) 129

(76.2–85.5) (51.1–62.8) (5.7–12.4) (3.0–8.4) (3.6–9.5) (2.4–7.5)

o0.001 o0.001 o0.001 o0.001 o0.01 0.03

Abbreviations: CI, confidence interval; F, female; M, male; T, total. Data are mean (s.d.) or age adjusted percent (95%CI). *On the basis of w2 test across all groups.**Comparing total across all groups.

hyperlipidaemia compared to European Caucasians and Afro-Caribbeans (Po0.05; Table 1). Secular trends in non-haemorrhagic stroke admissions

Between 1997 and 2005, there was a secular decline in admission rates of non-haemorrhagic stroke (Po0.001), from 1.6 per 1000 admissions in the period 1997–1999 to 0.95 in the period 2003–2005. Between ethnic groups, the magnitude of the secular decline in rates of non-haemorrhagic stroke admissions varied, where the decrease among European Caucasian and Afro-Caribbean populations was greater than that observed in the South Asian population (Po0.03). Secular trends in cardiovascular risk factors

Hypertension, peripheral vascular disease and hyperlipidaemia were all increased between secular time periods between 1997 and 2005 (Po0.001). Myocardial infarction was the only risk factor that showed a significant decrease in prevalence during the same periods for the total stroke population (Po0.001). Among South Asians, hypertension (P ¼ 0.026), peripheral vascular disease (P ¼ 0.019) and hyperlipidaemia (Po0.001) showed a significant increase during the period 1997–2005. During the same period, European Caucasians also showed a significant increase in the prevalence of hypertension (Po0.001), hyperlipidaemia (Po0.001) and peripheral vascular disease (P ¼ 0.003). In contrast, myocardial infarction was decreased in the European Caucasian population (Po0.001). In Afro-Caribbeans, atrial fibrillation (P ¼ 0.039) and hyperlipidaemia (P ¼ 0.007) showed a significant increase in the period 1997–2005 (Figure 1; Table 2). Duration of stay and in-hospital mortality

The average duration of hospital stay for the total population was 20.9 (s.d. 75.2) days. Afro-Caribbeans showed a significantly higher duration of hospital stay Journal of Human Hypertension

compared to that of European Caucasians and South Asians, 29.7 vs 20.9 and 21.63 days, Po0.001. On survival analysis, cumulative event-free survival over 30 days varied significantly by ethnicity, but significantly decreased in all groups (P ¼ 0.031). European Caucasians had the poorest survival between 1997 and 1999 (P ¼ 0.02), while South Asians had the poorest survival for the time periods 2000–2002 (P ¼ 0.29) and 2003–2005 (P ¼ 0.03; Table 2). On Cox regression analysis of 30-day mortality, incorporating age, gender and cardiovascular risk factors (myocardial infarction, peripheral vascular disease, hypertension, hyperlipidaemia, diabetes and atrial fibrillation), hypertension (odds ratio, 95% CI 1.51 (1.15–1.97), P ¼ 0.028) and age (1.02 (1.01–1.03), P ¼ 0.001) were independent predictors in the first time period. Between 2000 and 2002, hyperlipidaemia was a nonsignificant predictor of 30-day mortality (P ¼ 0.06). Between 2003 and 2005, South Asian ethnicity was an independent predictor of mortality vs European Caucasians (2.47 (1.10–5.54), P ¼ 0.03) and Afro-Caribbeans (2.50 (1.23–5.08), P ¼ 0.02).

Discussion Contrary to our null hypothesis, there was a decline in rates of hospital admissions with non-haemorrhagic strokes for the total population in recent years, which does correspond to an improvement in CVD risk profile as in myocardial infarction and diabetes for the total population. However, increased hypertension and hyperlipidaemia were seen over the studied period, but may be explained by how the identification of risk factors has improved among stroke patients, in line with a holistic approach in CVD risk management, specifically in South Asians.21 The data here give an insight into the changing CVD risk profile among ethnic groups. Hence, in South Asians, a CVD risk profile that includes more hyperlipidaemia is likely

Ethnicity, cardiovascular risk and stroke A Gunarathne et al 21 Hypertension

European Caucasian

90

Diabetes Myocardial Infarction

65

Hyperlipidaemia Atrial Fibrillation

40

p < 0.001

p = 0.97 p = 0.39

15

p < 0.001

90 p = 0.026

South Asian

Age Adjusted Percentage

p < 0.001

65 p = 0.66 40 p < 0.001 p = 0.19

15

p = 0.07

Afro Caribbean

90

p = 0.05

65

p = 0.17

40

p = 0.007 * p = 0.039 **

15

p = 0.18

1997-1999

2000-2002

2003-2005

Time Period

Figure 1 Cardiovascular risk profile by time period and ethnic groups. P-values using w2 analysis comparing 1997–1999 period vs 2003–2005 period are shown. Error bars show 95% confidence intervals. *P-value for hyperlipidaemia. **P-value for atrial fibrillation.

to explain why rates of stroke have not fallen to the same extent as those seen in other ethnic groups. While hyperlipidaemia was more common in all ethnic groups, the impact was more so in South Asians, and this may explain the greater 30-day mortality in this group. In recent years, there has been consensus recognition that South Asians is a high-risk group for CVD, and this may have promoted heightened CVD risk assessments specific to this population.1 Thus, it would be important to investigate the role of dyslipidaemia among South Asians in a community-based study, in view of its impact on stroke aetiology and pathophysiology. Using a hospital-

based approach, South Asian stroke survivors have been found to have significantly higher apolipoprotein B to AI ratios, and higher lipoprotein(a) levels when compared with ethnically matched healthy controls.22 Given that hyperlipidaemia is a commonly encountered risk factor in CVD, being present in more than 25% of the stroke patients, there is the need to determine the importance of dyslipidaemic management in the treatment and prevention of stroke in a multiethnic population. In two other reported South Asian stroke studies, 50% of the stroke patients had elevated cholesterol levels.23,24 In European Caucasians, the prevalence of hyperlipidaemia ranges from 28.7 (Dijon, France), 29.5 (Oxford, UK) to 32% (Manhattan study, USA).25 In the present study, the prevalence of hyperlipidaemia increased by fivefold in South Asians over the period 1997–2005. This raises the possibility of adverse transition of the CVD protective traits in hyperlipidaemia, which were a feature of early migrant populations, to a more atherogenic profile as described in the indigenous population.13 The majority of the other strokes studies in the United Kingdom have supported a similar decline in the prevalence of stroke18 but the pace of decline in stroke prevalence in South Asians was not significant, reaching a nadir across the three time periods. This is consistent with the adverse transition of CVD risk profile among South Asians shown here. This study also shows a declining trend in inhospital mortality rates in all three ethnic groups during the study period. This is well comparable to other reported studies across the world.26 In the United Kingdom, ethnic disparity in stroke mortality rates has been reported.29 However, highest case fatality rates in South Asians compared to all other ethnic groups in the current study further demonstrate the vulnerability of this group to the whole spectrum of CVD.12 There are several limitations to the study analyses, including the hospital-based retrospective registry nature, although stroke prevalence was indirectly determined by calculating proportional hospital admission rates, which is an accepted epidemiological method.27 Also, the reporting of cardiovascular comorbidities and stroke type is limited by the quality of information that is recorded within case note histories. Reported prevalence rates in this study are purely based on the hospital stroke admissions, where there could be an underestimation of the stroke prevalence in the general at-risk population in the community. However, hospital admission rates in this study are proportional to the prevalence rates reported by Stewart et al.28 based on the London stroke registry data and population-based stroke study reported by Hsu et al.29 in Leicestershire. The stable population structure observed among ethnic groups in the hospital catchment area over the studied period Journal of Human Hypertension


Table 2 Secular trends in demographics and cardiovascular risk profiles for patients admitted with non-haemorrhagic strokes between 1997 and 2005 1997–1999 (n ¼ 986)

2000–2002 (n ¼ 690)

2003–2005 (n ¼ 729)

P-value

Age, years mean (s.d.) Male Female

73.9 (10.9) 467 (47.4%) 519 (52.6% )

74.8 (11.0) 325 (47.1%) 365 (52.9%)

75.8 (11.1) 329 (45.1%) 400 (54.9%)

o0.01 0.02 0.53

Ethnic group breakdown, n (%) Caucasian men Caucasian women South Asian men South Asian women Afro-Caribbean men Afro-Caribbean women

319 403 85 60 63 56

(32.3) (40.9) (8.6) (6.1) (6.4) (5.7)

194 264 83 60 48 41

(28.1) (38.3) (12.0) (8.9) (7.0) (5.9)

218 311 75 57 36 32

(29.9) (42.7) (10.2) (7.8) (4.9) (4.4)

o0.001 o0.001 0.71 0.95 0.02 0.03

Hospital admission rate (95% CI) Total Caucasian South Asian Afro-Caribbean

1.60 1.60 1.19 2.4

(1.5–1.7) (1.5–1.7) (1.1–1.3) (12.2–2.6)

1.06 1.06 0.98 1.11

(1.0–1.1) (1.0–1.1) (0.9–1.1) (1.0–1.2)

0.95 0.99 0.86 0.86

(0.9–1.0) (0.9–1.0) (0.8–0.9) (0.8–1.0)

o0.001 o0.001 0.03 o0.001

Risk factors, n (%) Hypertension Myocardial infarction Peripheral vascular disease Hyperlipidaemia Atrial fibrillation Diabetes Ischaemic heart disease Valvular heart Disease Transient ischaemic attacks

539 93 50 33 267 367 228 23 72

(54.7) (9.4) (5.0) (3.3) (27.1) (37.2) (23.1) (2.3) (5.7)

446 38 31 61 177 249 156 14 73

(64.6) (5.5) (4.5) (8.8) (25.7) (36.1) (22.6) (2.0) (5.8)

515 26 71 112 211 269 185 15 96

(70.6) (3.6) (9.7) (15.4) (28.9) (36.9) (25.3) (2.1) (7.7)

o0.001 o0.001 o0.001 o0.001 0.38 0.89 0.41 0.89 0.15

17.9 (1.5) 273 (226–320)

o0.001 0.03

Duration of stay (days) mean (s.d.) Stroke mortality rate (95% CI)

24.8 (1.6) 319 (275–364)

20.7 (1.6) 303 (257–349)

Abbreviation: CI, confidence interval. P-value using w2 analysis, except age of admission where analysis of variance (ANOVA) is used. Data are n (%) or mean (s.d.) and rates are per 1000 patients per year (95% CI). Statistical significance was accepted at Po0.05.

legitimates the comparison of prevalence rates between the ethnic groups. Although there has been no change of the ICD 10th revision during the period of our study, threshold levels for diseases like hypertension have changed. This could result in an overestimation of the prevalence rates when analysing and interpreting secular trends. Similarly, there could also be an overdiagnosis of strokes on admissions, specifically in older patients, who are admitted with many other comorbidities. Nonetheless, it may not have contributed to the determination of ethnic differences, as one would assume that changes would apply equally to all cohorts. In conclusion, cardiovascular risk profiles among patients admitted with non-haemorrhagic stroke have changed over the last decade. In particular, hyperlipidaemia has increased, especially among South Asians, where it showed a fivefold increase. The reduced decline in stroke prevalence and 30-day survival of stroke in South Asians in recent years warrants further investigation and highlights the importance of a targeted health-care approach in the migrant ethnic minorities. Journal of Human Hypertension

What is known about this topic K The burden of ischaemic stroke is more prominent among UK migrant communities. K This may relate to the impact of its determinants among these ethnic minorities, most notably, hypertension and diabetes mellitus, which are frequently manifest among South Asian and Afro-Caribbean ethnic groups. What this study adds Cardiovascular risk profiles among patients admitted with non-haemorrhagic stroke have changed over the last decade. K In particular, hyperlipidaemia has increased, especially among South Asians, where it showed a fivefold increase. K The reduced decline in stroke prevalence and 30-day survival of stroke in South Asians in recent years warrants further investigation. K

References 1 Lip GY, Barnett AH, Bradbury A, Cappuccio FP, Gill PS, Hughes E et al. Ethnicity and cardiovascular disease prevention in the United Kingdom: a practical approach to management. J Hum Hypertens 2007; 21: 183–211. 2 Rothwell PM, Coull AJ, Silver LE, Fairhead JF, Giles MF, Lovelock CE, et al., Oxford Vascular Study.


3 4

5 6

7 8 9

10 11 12

13

14

15

16

Population-based study of event-rate, incidence, case fatality, and mortality for all acute vascular events in all arterial territories (Oxford Vascular Study). Lancet 2005; 366: 1773–1783. Wild S, McKeigue P. Cross sectional analysis of mortality by country of birth in England and Wales, 1970–92. BMJ 1997; 314: 705–710. Raleigh VS, Kiri V, Balarajan R. Variations in mortality from diabetes mellitus, hypertension and renal disease in England and Wales by country of birth. Health Trends 1997; 28: 122–127. Khattar RS, Swales JD, Senior R, Lahiri A. Racial variation in cardiovascular morbidity and mortality in essential hypertension. Heart 2000; 83: 267–271. Lane DA, Lip GY, Beevers DG. Ethnic differences in cardiovascular and all-cause mortality in Birmingham, England: the Birmingham Factory Screening Project. J Hypertens 2005; 23: 1347–1353. Lane DA, Lip GYH. Ethnic differences in hypertension and blood pressure control in the UK. Q J Med 2001; 94: 391–396. Cappuccino FP, Barbato A, Kerry SM. Hypertension, diabetes and cardiovascular risk in ethnic minorities in the UK. Br J Diabetes Vasc Dis 2003; 3: 286–293. Schulz UG, Rothwell PM. Differences in vascular risk factors between etiological subtypes of ischemic stroke: importance of population-based studies. Stroke 2003; 34: 2050–2059. Gillum RF. Epidemiology of stroke in Hispanic Americans. Stroke 1995; 26: 1707–1712. Gillum RF. Risk factors for stroke in blacks: a critical review. Am J Epidemiol 1999; 150: 1266–1274. Cruickshank JK, Alleyne SA. Black West Indian and matched white diabetics in Britain compared with diabetics in Jamaica: body mass, blood pressure, and vascular disease. Diabetes Care 1987; 2: 170–179. 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: 405–409. Patel JV, Vyas A, Cruickshank JK, Prabhakaran D, Hughes E, Reddy KS et al. Impact of migration on coronary heart disease risk factors: comparison of Gujaratis in Britain and their contemporaries in villages of origin in India. Atherosclerosis 2006; 185: 297–306. Cruickshank JK, Cooper J, Burnett M, MacDuff J, Drubra U. Ethnic differences in fasting plasma Cpeptide and insulin in relation to glucose tolerance and blood pressure. Lancet 1991; 338: 842–847. Lyratzopoulos G, McElduff P, Heller RF, Hanily M, Lewis PS. Comparative levels and time trends in blood pressure, total cholesterol, body mass index and smoking among Caucasian and South-Asian partici-

17 18

19 20

21

22

23

24

25

26

27 28

29

pants of a UK primary-care based cardiovascular risk factor screening programme. BMC Public Health 2005; 5: 125. George LM, Maddock K. Accuracy of hospital activity notifications for infectious diseases. BMJ 1979; 1: 1332–1339. NHS Strategic Tracing Service. Better information for health where and when it’s needed—NHS connecting or health. http://www.connectingforhealth.nhs.uk,accessed on 2/8/2006. Hatona S. Experiences from a multicenter stroke register: a preliminary report. Bull World Health Organ 1976; 54: 541–553. National Institute of Clinical Excellence (NICE). National Institute of Clinical Excellence (NICE) guidelines for cardiovascular disease. http://guidance.nice.org.uk/topic/cardiovascular,accessed on 2/8/2006. British Cardiac Society; British Hypertension Society; Diabetes UK; HEART UK; Primary Care Cardiovascular Society; Stroke Association. JBS 2: Joint British Societies guidelines on prevention of cardiovascular disease in clinical practice. Heart 2005; 91: 1–52. Sharobeem KM, Patel JV, Ritch AES, Gill PS, Lip GYH, Hughes EA. Elevated lipoprotein(a) and apolipoprotein B to AI ratio in South Asian patients with ischaemic stroke. Int J Clin Pract (in press). Deleu D, Hamad AA, Kamram S, El Siddig A, Al Hail H, Hamdy SM. Ethnic variations in risk factor profile, pattern and recurrence of non-cardioembolic ischemic stroke. Arch Med Res 2006; 37: 655–662. 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). Indian Heart J 2000; 52: S35–S43. Sacco RL, Boden-Albala B, Abel G, Lin IF, Elkind M, Hauser WA et al. Race-ethnic disparities in the impact of stroke risk factors: the northern Manhattan stroke study. Stroke 2001; 32: 1725–1731. Howard G, Howard VJ, Katholi C, Oli MK, Huston S. Decline in US stroke mortality: an analysis of temporal patterns by sex, race, and geographic region. Stroke 2001; 32: 2213–2220. WHO. WHO STEPwise approach to surveillance. http://www.who.int/chp/steps,accessed on 2/8/2006. Stewart JA, Dundas R, Howard RS, Rudd AG, Wolfe CD. Ethnic differences in incidence of stroke: prospective study with stroke register. BMJ 1999; 318: 967–971. Hsu RT, Ardron ME, Brooks W, Cherry D, Taub NA, Botha JL. The 1996 Leicestershire Community Stroke & Ethnicity Study: differences and similarities between South Asian and white strokes. Int J Epidemiol 1999; 28: 853–858.

Journal of Human Hypertension