Abstract: Objective. To describe gender-specific long-term outcome and initiation of secondary preventive medication among patients with acute myocardial infarction (AMI). Design. Observational cohort study. Setting. Nationwide registries. Patients. We included 18,279 patients: 6364 women (35%) and 11,915 men (65%), admitted with AMI (median age, 67 years; range, 30-90 years) surviving for at least 2 months. Interventions. According to sex, patients were stratified by invasive treatment strategy: (1) revascularized; (2) examined with coronary angiography (CAG) but not revascularized; and (3) not examined with CAG. Main outcome measures. All-cause mortality and readmission with AMI. Initiation of secondary preventive medication. Results. Of 18,279 patients with a first AMI who survived 2 months, 1857 women (29%) and 1756 men (15%) were not examined with CAG (P<.001), 1295 women (20%) and 1563 men (13%) were examined but not revascularized (P<.001), and 3212 women (51%) and 8596 men (72%) were revascularized (P<.001). Not being examined with CAG after AMI was associated with a threefold increase in risk of death and, importantly, a 50% increase in the risk of a recurrent AMI compared with patients who were revascularized. Among patients who were revascularized, 85-92% initiated recommended secondary preventive medication compared to 46-71% in patients not examined with CAG (P<.001). Initiation of secondary preventive medication was higher in men (81-84%) than in women (73-79%; P<.001), which could be ascribed to the differences in invasive strategy. Conclusions. In both sexes, those who were not examined had a highly increased risk of both recurrent AMI and death. Moreover, initiation of secondary preventive medication was closely related to the choice of invasive strategy disfavoring the women.
J INVASIVE CARDIOL 2012;24(1):19-24
Key words: acute coronary syndrome, gender, coronary angiography, coronary intervention (PCI), delivery of care
The treatment of acute myocardial infarction (AMI) has evolved rapidly over the last decades. An invasive strategy for AMI patients has been shown to improve cardiovascular morbidity and mortality.
In Denmark, following the results of the DANAMI 2 trial,1 treatment with primary percutaneous intervention (primary PCI) has been implemented for all patients with ST-segment elevation myocardial infarction (STEMI), and an early invasive strategy for non-STEMI patients is recommended according to the results of the FRISC II trial, the RITA 3 trial, and the TACTICS-TIMI 18 trial.2-4 Current international guidelines recommend primary PCI for patients with STEMI if a PCI facility is reachable within 120 minutes.5 For patients with non-STEMI, invasive examination and treatment is recommended within 72 hours.6 Thus, coronary angiography (CAG) is recommended for all AMI patients, but the timing of the diagnostic and therapeutic process differs for those with and without STEMI. In Denmark, more than 20% of patients admitted with AMI who survive the initial day of admission are not examined with coronary angiography;7 these patients are often older and have more comorbidity. Medications for long-term secondary prevention — antiplatelet drugs, beta-blockers, and lipid-lowering drugs (statins) — have proven to be highly effective in reducing the risk of cardiovascular morbidity and mortality after AMI.8-10 Guidelines recommend secondary preventive medication for all AMI patients, provided there are no contraindications.5,6
Women are at higher risk of adverse outcomes than men, which has primarily been ascribed to women’s higher age and increased comorbidity.11,12 Additionally, women receive less secondary preventive medication.13-18 Gender differences in the invasive treatment of patients admitted with AMI have been shown in several studies.7,13,19 The significance of this is controversial; coronary anatomy differs between sexes, and this can largely explain differences in revascularization rate among those who undergo CAG. However, the choice of referral to invasive examination also differs by gender,7 and this may have implications in terms of poorer medical treatment, rehabilitation, and follow-up.
The aims of the present study were: (1) to determine the long-term prognosis associated with the invasive strategy for men and for women; (2) to investigate the relation between the invasive strategy and gender differences in implementation of secondary preventive medication.
The inclusion period was 1 January 2005 through 31 December 2007. Data were collected from four registers: the Danish National Patient Registry, which holds information on all admissions to all Danish hospitals since 1978;20 the Danish Civil Registration System, where all persons residing in Denmark are registered from birth or time of immigration; the Danish Heart Registry,21 which includes data from all coronary angiographies and revascularizations by PCI or coronary artery bypass grafting (CABG) at Danish hospitals since the year 2000; and the Danish Registry of Medicinal Product Statistics, where all redeemed prescriptions in Denmark have been recorded since 1995.22 All Danish residents have a unique and permanent personal civil registration number, which allows linkage between nationwide registries on an individual level.23
Study population. We identified all patients aged 30-90 years admitted with AMI as the primary diagnosis (International Classification of Diseases [ICD]-10, codes I21-I22). Information on admissions and comorbidity was obtained from the Danish National Patient Registry. Each admission is registered by the final discharge diagnoses: one primary diagnosis and, if appropriate, one or more secondary diagnosis according to the ICD — before 1994 the 8th revision (ICD-8) and after 1994 the 10th revision (ICD-10). The definition of AMI in Denmark follows the guidelines of the European Society of Cardiology.5,6 To study a first admission with AMI, we excluded all patients with a previous diagnosis of AMI from 1978-2004. All patients discharged on the day of admission were excluded to ensure the diagnosis was based on reliable observations and a sufficient number of blood tests.7,24 In a validity study of AMI recorded in the registry, the sensitivity and specificity were 90% and 95%, respectively.25,26 Another study found a positive predictive value of 92.4% for the diagnosis of AMI using the same inclusion criteria as in our study but without applying our strict exclusion criteria.27
To study long-term prognosis and risk of reinfarction in relation to the selected invasive strategy, we excluded all patients who died within 2 months. Information on the patients’ vital statuses (alive or date of death) was obtained from the Danish Civil Registration System.
Comorbidity. Primary and secondary diagnoses, both at the index admission and at admissions up to 1 year before the index admission, were used to define comorbidity.28 Diagnoses of congestive heart failure, cardiogenic shock, arrhythmia, and pulmonary edema gave indications of the severity of the heart disease, while diagnoses of malignancy, diabetes with complications, cerebrovascular disease, acute renal failure, and chronic renal failure indicated comorbidity. This method is an extension of the Ontario myocardial infarction mortality prediction rule29 with translation to ICD-10 codes.30 Previous revascularization within 5 years of the index event was recorded for each patient, as this might have influenced the decision to initiate an invasive examination.
Medication. Information on medication before and after the event was obtained from the Danish Registry of Medicinal Product Statistics. In Denmark, all cardiovascular pharmaceuticals except aspirin require a prescription. Owing to partial reimbursement of drug expenses by the health care system, pharmacies in Denmark are required to register all dispensed prescriptions in the Danish Registry of Medicinal Product Statistics. This ensures a highly accurate registry.22 The drugs are classified according to the international Anatomical Therapeutical Chemical (ATC) system. Prior medication was defined as a prescription for the drug within the 6 months prior to the admission date. Eight groups of medical therapy were defined: beta-blockers (ATC-code C07); loop diuretics (C03EB, C03C); renin-system blockers (C09); lipid-lowering drugs (C10); aspirin (B01AC06, N02BA01); calcium-channel blockers (C08); anti-diabetics (A10); and chronic obstructive pulmonary disease medicine (R03).
Invasive procedures. Information on coronary angiography as well as PCI and CABG in patients examined with coronary angiography was retrieved from the Danish Heart Registry.21 To determine the most appropriate observation period from AMI to revascularization, we analyzed proportions revascularized within 30, 60, 90, 120, 150, and 180 days. We found that the majority of patients were revascularized within 60 days; this was used as a cut-off value. The procedures and examination results were validated by comparing a stratified random sample of 200 procedures in the registry (50 coronary angiographies, 50 PCI, 50 CABG, and 50 valve replacements) with local registries and written patient records. All 200 procedures were confirmed and the results of the 50 coronary angiographies accorded with the patients’ records.
Outcomes. The prognostic outcomes were clinical endpoints of all-cause mortality and recurrence of AMI from day 60 of index AMI and up to 1600 days. Initiation of secondary preventive medication was evaluated as a prescription redeemed within 60 days of index AMI for aspirin, beta-blockers, clopidogrel (B01AC04), and lipid-lowering drugs. Other ATC codes are listed above.
Statistical analysis. Continuous variables are presented as mean ± standard deviation (SD) for characteristics with appropriately near-symmetrical distributions or as median with interquartile range (IQR). Discrete data are presented as frequencies and percentages. Sex differences in the baseline characteristics were evaluated using Pearson’s chi-square test for categorical variables and student’s t-test for comparison of continuous data if normally distributed, otherwise non-parametric Mann-Whitney U-test was used.
Mortality after the index event is illustrated by plots of the cumulative hazard function adjusted for confounding variables. Differences in mortality rate were estimated using Cox proportional hazards regression. As differences in mortality could potentially influence the rate of admission for recurrent AMI, we analyzed the rate of recurrent AMI in a competing risks model. Cumulative incidence curves were plotted for recurrent AMI adjusted for confounding variables. Differences in recurrent AMI rates were estimated using competing-risks regression based on a proportional sub-hazards model.31 Model assumptions — linearity, proportional hazards, and interactions — were tested and found valid unless otherwise indicated. All hypothesis tests had a .05 significance level and were two-sided. Differences in the initiation of secondary preventive medication were analyzed using logistic regression analysis. All data handling was done with SAS version 9.14 (SAS Institute Inc.), and all analyses were performed with STATA version 11 (StataCorp).
Ethics. The Danish National Board of Health, the Danish Data Protection Agency, and the Board of the Danish Heart Registry all approved the project, which was conducted in accordance with the current rules of ethics and legislature. Data were encrypted, meaning that no individual patient could be identified. Register-based studies do not require ethical approval in Denmark.
During the study period, a total of 20,702 patients were admitted with first-time AMI, of whom 18,279 were discharged and alive after 60 days. Thus, 18,279 patients were included in the study: 11,915 men (65%) and 6364 women (35%). In all, 8596 men (72%) and 3212 women (50.5%) were revascularized (P<.001). Of those examined with CAG and not revascularized, 1563 were men (13%) and 1295 were women (20%) (P<.001). Finally, there were 1857 women (29%) and 1756 men (15%) who were not examined with CAG (P<.001). Thus, proportionately twice as many women as men, almost a third of all women, were not undergoing an invasive strategy after AMI.
Baseline characteristics. Baseline characteristics of the population stratified by sex and treatment strategy are outlined in Table 1. Women were generally older and had more comorbid conditions than men did, but these differences were attenuated when stratified by treatment strategy, although significant differences still existed. The same pattern was seen for concomitant medical treatment where the overall pattern of significant differences persisted only among those revascularized. There was a total of 57,459 years of follow-up, with a mean follow-up of 1148 days (3.1 years) and a maximum of 1833 days.
All-cause mortality. A total of 3505 patients died: 1953 men and 1552 women. Figure 1A illustrates the adjusted cumulative hazard stratified by treatment strategy for men and women. We found that patients not examined with coronary angiography had higher mortality than those examined. Figure 2A shows that men and women not examined with CAG had an equally increased risk of death (hazard ratio [HR] of 3.40 and 3.36, respectively, adjusted for age and comorbidity, P=.80 for interaction) when compared with revascularized patients of the same sex.
Recurrent AMI. In all, there were 1397 recurrent AMIs: 877 in men and 520 in women. Figure 1B shows the cumulative incidence by treatment strategy adjusted for age and comorbidity for men and women. The results are similar to those of all-cause mortality, albeit the absolute numbers are lower. Figure 2B shows that HRs were higher among those not revascularized, with the same excess risk for both men and women (HR 1.68 for non-examined men compared with revascularized men and HR 1.49 for non-examined women compared with revascularized women; P=.38 for interaction).
Secondary preventive medical treatment. There was a clear difference between the invasive strategy groups regarding secondary preventive medical treatment. The largest proportion of patients redeeming prescriptions was among those who were revascularized (84-92%), with a lower proportion among those invasively examined (63-79%) and with the lowest proportion among patients who were not examined (49-70%) (Table 1).
The proportions of patients redeeming a prescription for secondary preventive medication differed overall between men and women. When stratifying by treatment strategy, these differences were somewhat attenuated and larger absolute differences persisted only in the non-examined group where women, compared with men, were less treated with beta-blockers and lipid-lowering drugs. Men were more likely to initiate treatment with beta-blockers (odds ratio [OR], 1.24; 95% confidence interval [CI], 1.14-1.35), lipid-lowering drugs (OR, 1.22; 95% CI, 1.12-1.32), and clopidogrel (OR, 1.25; 95% CI, 1.16-1.35) and the same trend was seen with aspirin (OR, 1.07; 95% CI, 0.99-1.16), although not statistically significant (Figure 3).
The main finding of this study was that patients with their first AMI who were not revascularized were at increased risk of recurrent AMI and death after adjustment of age and comorbidity. This underlines the importance of initiating secondary preventive medication in all patients as well as the importance of examining patients with AMI with CAG to identify those who could benefit from revascularization.
A consistent pattern in the literature shows a greater mortality among women than men with AMI.12,32,33 and that this difference can largely be ascribed to age and comorbidity. This is in line with the results in our population. Proportionately twice as many women as men, almost a third of all women, were not examined with CAG after AMI. We showed that those not examined had a highly increased risk of both death and, most importantly, recurrent AMI in both sexes. This is important because it highlights the fact that although patients who are not examined are older and have more comorbid conditions and thus have a high mortality rate; they are also at a high risk of experiencing a recurrent myocardial infarction.
Initiation of secondary preventive medication was closely related to the choice of invasive strategy, with more revascularized patients initiating all four drugs than those not revascularized, and with those not examined having the lowest proportion of initiation. Findings have indicated that initiation of secondary preventive medication is associated with the chosen invasive strategy34,35 — our findings concur with this — and optimal medical therapy has been related to better outcomes independently of revascularization strategy.34,35 After adjusting for age and comorbidity, we found that gender-specific initiation of secondary preventive medication was more likely among men than women. We demonstrated that this is especially related to the fact that proportionately twice as many women as men were among those not examined with CAG. Patients offered an invasive strategy with or without intervention have had more contact with the health care system due to the procedures —some have had cardiac surgery and others have had stents implanted into their coronary arteries. Focus on the preservation of the technically good results of the PCI or CABG and on prevention of possible stent thrombosis may explain the increased initiation of secondary preventive medical treatment among the revascularized patients. Because of the differences in the invasive strategy of men and women, women are less likely to initiate secondary preventive medication. This could have implications for the outcome of women, since the non-revascularized and non-examined patients are at increased risk of experiencing recurrent AMI. Patients not starting treatment soon after discharge have a low probability of starting treatment later.18 Consequently, it is important to focus on these patients in the early phase.
Strengths and limitations of the study. The contemporary data and the completeness of the registries give valuable information on the long-term prognosis of AMI patients depending on the chosen invasive treatment strategy in the present clinical setting, and the findings point toward areas needing close attention. The data covered the entire population of Denmark, independent of race, socioeconomic status, age, or participation in health insurance programs, and included citizens both in and outside the labor market. Therefore, the risk of selection bias was minimized. Following a directive issued by the Danish Board of Health, reporting to the Danish Heart Registry is mandatory for all hospitals, both public and private, performing the procedures, which strengthens the completeness of the data. Patients had a 1-day stay at minimum and were discharged with a primary diagnosis of AMI, thus adding to the validity of the diagnosis. The main limitation was inherent in the observational nature of the study. Other limitations of our analysis that warrant consideration were the differences in the clinical presentation of AMI, on which we had little information. We had no information on key clinical variables, such as electrocardiographic changes on presentation and elevation in biomarkers, thus we were unable to specify on which of these parameters the diagnosis was made; furthermore, we were unable to distinguish between STEMI and non-STEMI patients. Lastly, the detailed causes for choosing not to perform coronary angiography and not to revacularize the patients following coronary angiography have not been evaluated. Accordingly, the reasons for the increased morbidity and mortality in those patients need further study.
Conclusions and implications. Not undergoing CAG after AMI is associated with a threefold increase in risk of death in both men and women and, importantly, also a 50% increase in the risk of a recurrent AMI compared with those patients who are revascularized. Therefore, initiation of secondary preventive medication is important in all patients diagnosed with AMI; nevertheless, initiation was much less intensive among those not offered the invasive treatment strategy. Thus, focus on both choice of invasive strategy and secondary preventive medication in all patients discharged with AMI is still important, especially in women, since 29% do not undergo invasive treatment.
Acknowledgment. The authors thank the contributors to the Danish Heart Registry for the work put into its continuous operation.
- Andersen HR, Nielsen TT, Rasmussen K, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med. 2003;349(8):733-742.
- Invasive compared with non-invasive treatment in unstable coronary-artery disease: FRISC II prospective randomised multicentre study. FRagmin and Fast Revascularisation during InStability in Coronary artery disease Investigators. Lancet. 1999;354(9180):708-715.
- Fox KAA, Poole-Wilson PA, Henderson RA, et al. Interventional versus conservative treatment for patients with unstable angina or non-ST elevation myocardial infarction: the British Heart Foundation RITA 3 randomised trial. Lancet. 2002;360(9335):743.
- Mahoney EM, Jurkovitz CT, Chu H, et al. Cost and cost-effectiveness of an early invasive vs conservative strategy for the treatment of unstable angina and non-ST segment elevation myocardial infarction. JAMA. 2002;288(15):1851-1858.
- Van de Werf F, Bax J, Betriu A, et al. Management of acute myocardial infarction in patients presenting with persistent ST-segment elevation: the Task Force on the Management of ST-Segment Elevation Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J. 2008;29(23):2909-2945.
- Bassand JP, Hamm CW, Ardissino D, et al. Guidelines for the diagnosis and treatment of non-ST segment elevation acute coronary syndromes. Eur Heart J. 2007;28(13):1598-1660.
- Hvelplund A, Galatius S, Madsen M, et al. Women with acute coronary syndrome are less invasively examined and subsequently less treated than men. Eur Heart J. 2010;31(6):684-690.
- Freemantle N, Cleland J, Young P, et al. beta Blockade after myocardial infarction: systematic review and meta regression analysis. BMJ. 1999;318(7200):1730-1737.
- Collaborative meta-analysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ. 2002;324(7329):71-86.
- Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360(9326):7-22.
- Berger JS, Elliott L, Gallup D, et al. Sex differences in mortality following acute coronary syndromes. JAMA. 2009;302(8):874-882.
- Halvorsen S, Eritsland J, Abdelnoor M, et al. Gender differences in management and outcome of acute myocardial infarctions treated in 2006-2007. Cardiology. 2009;114(2):83-88.
- Nguyen JT, Berger AK, Duval S, et al. Gender disparity in cardiac procedures and medication use for acute myocardial infarction. Am Heart J. 2008;155(5):862-868.
- Dey S, Flather MD, Devlin GP, et al. Sex-related differences in the presentation, treatment and outcomes among patients with acute coronary syndromes. The Global Registry of Acute Coronary Events. Heart. 2009;95(1):20-26. Epub 2008 May 7.
- Reina A, Colmenero M, Aguayo de Hoyos E, et al. Gender differences in management and outcome of patients with acute myocardial infarction. Int J Cardiol. 2007;116(3):389-395.
- Radovanovic D, Erne P, Urban P, Bertel O, Rickli H, Gaspoz JM; AMIS Plus Investigators. Gender differences in management and outcomes in patients with acute coronary syndromes: results on 20,290 patients from the AMIS Plus Registry. Heart. 2007;93(11):1369-1375.
- Milcent C, Dormont B, Durand-Zaleski I, Steg PG. Gender differences in hospital mortality and use of percutaneous coronary intervention in acute myocardial infarction: microsimulation analysis of the 1999 nationwide French hospitals database. Circulation. 2007;115(7):833-839.
- Gislason GH, Rasmussen JN, Abildstrom SZ, et al. Long-term compliance with beta-blockers, angiotensin-converting enzyme inhibitors, and statins after acute myocardial infarction. Eur Heart J. 2006;27(10):1153-1158.
- Heer T, Schiele R, Schneider S, et al. Gender differences in acute myocardial infarction in the era of reperfusion (the MITRA registry). Am J Cardiol. 2002;89(5):511-517.
- Andersen TF, Madsen M, Jørgensen J, Mellemkjoer L, Olsen JH. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull. 1999;46(3):263-268.
- Abildstrøm SZ, Kruse M, Rasmussen S, Madsen JK, Nielsen PH, Madsen M; Danish Heart Registry. [The Danish Heart Registry — a clinical database]. Ugeskr Laeger. 2008;170(7):532-536.
- Gaist D, Sorensen HT, Hallas J. The Danish prescription registries. Dan Med Bull. 1997;44(4):445-448.
- Pedersen CB, Gotzsche H, Moller JO, et al. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull. 2006;53(4):441-449.
- Rasmussen JN, Rasmussen S, Gislason GH, et al. Persistent socio-economic differences in revascularization after acute myocardial infarction despite a universal health care system — a Danish study. Cardiovasc Drugs Ther. 2007;21(6):449-457.
- Madsen M, Davidsen M, Rasmussen S, Abildstrom SZ, Osler M. The validity of the diagnosis of acute myocardial infarction in routine statistics: a comparison of mortality and hospital discharge data with the Danish MONICA registry. J Clin Epidemiol. 2003;56(2):124-130.
- Nielsen KM, Foldspang A, Larsen ML, Gerdes LU, Rasmussen S, Faergeman O. Estimating the incidence of the acute coronary syndrome: data from a Danish cohort of 138 290 persons. EurJ Cardiovasc Prev Rehabil. 2007;14(5):608-614.
- Joensen AM, Jensen MK, Overvad K, et al. Predictive values of acute coronary syndrome discharge diagnoses differed in the Danish National Patient Registry. J Clin Epidemiol. 2009;62(2):188-194.
- Rasmussen S, Zwisler AD, Abildstrom SZ, Madsen JK, Madsen M. Hospital variation in mortality after first acute myocardial infarction in Denmark from 1995 to 2002: lower short-term and 1-year mortality in high-volume and specialized hospitals. Med Care. 2005;43(10):970-978.
- Tu JV, Austin PC, Walld R, et al. Development and validation of the Ontario acute myocardial infarction mortality prediction rules. J Am Coll Cardiol. 2001;37(4):992-997.
- So L, Evans D, Quan H. ICD-10 coding algorithms for defining comorbidities of acute myocardial infarction. BMC Health Serv Res. 2006;6:161.
- Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496-509.
- MacIntyre K, Stewart S, Capewell S, et al. Gender and survival: a population-based study of 201,114 men and women following a first acute myocardial infarction. J Am Coll Cardiol. 2001;38(3):729-735.
- Chang WC, Kaul P, Westerhout CM, et al. Impact of sex on long-term mortality from acute myocardial infarction vs unstable angina. Arch Intern Med. 2003;163(20):2476-2484.
- Lee JH, Yang DH, Park HS, et al. Suboptimal use of evidence-based medical therapy in patients with acute myocardial infarction from the Korea Acute Myocardial Infarction Registry: prescription rate, predictors, and prognostic value. Am Heart J. 2010;159(6):1012-1019.
- Yan AT, Yan RT, Tan M, et al. Optimal medical therapy at discharge in patients with acute coronary syndromes: temporal changes, characteristics, and 1-year outcome. Am Heart J. 2007;154(6):1108-1115.
From the aNational Institute of Public Health, University of Southern Denmark, Copenhagen, bDepartment of Cardiology, Copenhagen University Hospital Gentofte, Copenhagen, cDanish Heart Registry, www.dhreg.dk, dDepartment of Public Health, University of Copenhagen, Copenhagen, eDepartment of Cardiology, Aalborg Hospital, Aarhus University Hospital, Denmark, fDepartment of Cardiology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, gDepartment of Cardiothoracic Surgery, Odense University Hospital, Odense, hDepartment of Cardiothoracic Surgery, Skejby University Hospital, Aarhus, iDepartment of Cardiology, Copenhagen University Hospital Bispebjerg, Copenhagen.
Funding: This work was supported by the Danish Heart Foundation.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein. Manuscript submitted July 6, 2011, provisional acceptance given September 7, 2011, final version accepted September 27, 2011.
Address for correspondence: Anders Hvelplund, National Institute of Public Health, University of Southern Denmark, Øster Farimagsgade 5A, DK1399, Copenhagen K, Denmark. Email: firstname.lastname@example.org