HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Frost, L. Articles by Vestergaard, P. Search for Related Content PubMed PubMed Citation Articles by Frost, L. Articles by Vestergaard, P. Agricola Articles by Frost, L. Articles by Vestergaard, P.

Caffeine and risk of atrial fibrillation or flutter: the Danish Diet, Cancer, and Health Study^1,2,3

¹ From the Departments of Cardiology (LF) and Endocrinology and Metabolism (PV), Aarhus Sygehus, Aarhus University Hospital, Aarhus, Denmark.

² Supported by a senior research fellowship from the Danish Medical Research Council (grant no. 22-02-0582; to LF).

³ Reprints not available. Address correspondence to L Frost, Department of Cardiology, Aarhus Sygehus, Aarhus University Hospital, DK-8000 Aarhus C, Denmark. E-mail: Lars.Frost{at}as.aaa.dk.

See corresponding editorial on page 539.

	ABSTRACT

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Background: It is not known whether the consumption of caffeine is associated with excess risk of atrial fibrillation.

Objective: We evaluated the risk of atrial fibrillation or flutter in association with daily consumption of caffeine from coffee, tea, cola, cocoa, and chocolate.

Design: We prospectively examined the association between the amount of caffeine consumed per day and the risk of atrial fibrillation or flutter among 47 949 participants ( age: 56 y) in the Danish Diet, Cancer, and Health Study. Subjects were followed in the Danish National Registry of Patients and in the Danish Civil Registration System. The consumption of caffeine was analyzed by quintiles with Cox proportional-hazard models.

Results: During follow-up (: 5.7 y), atrial fibrillation or flutter developed in 555 subjects (373 men and 182 women). When the lowest quintile of caffeine consumption was used as a reference, the adjusted hazard ratios (95% CIs) in quintiles 2, 3, 4, and 5 were 1.12 (0.87, 1.44), 0.85 (0.65, 1.12), 0.92 (0.71, 1.20), and 0.91 (0.70, 1.19), respectively.

Conclusion: Consumption of caffeine was not associated with risk of atrial fibrillation or flutter.

Key Words: Arrhythmia • coffee • epidemiology

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Caffeine is a methylxanthine that is clinically similar to theophylline, and sympathomimetic effects due to circulating catecholamines cause the cardiac manifestations of caffeine overdose (1, 2). Caffeine toxicity by self-intended poisoning produces tachyarrhythmias such as supraventricular tachycardia, atrial fibrillation, ventricular tachycardia, and ventricular fibrillation (2-5).

The risk of cardiac arrhythmias associated with customary daily consumption of caffeine from coffee, tea, cola, cocoa, and chocolate does not seem to be increased at the ventricular level (6), but more information is needed about what happens at the supraventricular level (ie, the atrium).

The literature on caffeine and the risk of supraventricular arrhythmias is sparse. A study in 38 healthy subjects exposed to 1 mg caffeine/kg body wt after a 72-h period of caffeine abstention did not find that excess supraventricular arrhythmias were caused by exposure to caffeine (7). Conversely, atrial tachyarrhythmias including atrial fibrillation and flutter were produced by the intravenous administration of caffeine at a dose ranging from 1 to 5.0 mg caffeine/kg body wt in an experimental study in dogs (8). In the Multifactor Primary Prevention Study begun in Göteborg, Sweden, in 1970, consumption of 1–4 cups of coffee per day was associated with an age-adjusted odds ratio of atrial fibrillation of 1.24 (95% CI: 1.00, 1.54) during follow-up, whereas drinking >4 cups of coffee was not associated with a risk of atrial fibrillation (9). The purpose of the current study was to examine the association between the amount of caffeine consumed per day, and the risk of atrial fibrillation or flutter in the Danish Diet, Cancer, and Health Study.

	SUBJECTS AND METHODS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Study population
The Danish Diet, Cancer, and Health study is a prospective cohort study primarily focused on the role of diet in cancer risk, but which has the potential for studying other diseases as well. The study design has been described in detail elsewhere (10, 11).

From December 1993 through May 1997, subjects aged 50–64 y (n = 80 996 M, 79 729 F) were invited to participate in the study; 27 177 men and 29 876 women accepted the invitation. Eligible cohort members were born in Denmark, were living in the Copenhagen and Aarhus areas, and had no previous cancer diagnosis in the Danish Cancer Registry. The baseline data were linked to the Danish Cancer Registry and other population-based registries, including the Danish National Registry of Patients and the Danish Civil Registration System, by using the civil registry number, which is a unique number given to everyone living with an address in Denmark since 1968. The Civil Registration System has electronic records of all changes in vital status for the Danish population since 1968, including change in address, date of emigration, and date of death. The Danish National Registry of Patients, established in 1977, records 99.4% of all discharges from nonpsychiatric hospitals in Denmark (12). The data include the civil registry number, dates of admission and discharge, surgical procedures performed, and 20 discharge diagnoses per discharge, which were classified until 1993 according to the Danish version of the International Classification of Diseases, 8th revision (ICD-8; 13) and thereafter have been classified according to the Danish national version of the 10th revision (ICD-10; 14). In each case, the physician who discharged the patient coded all discharge diagnoses. To study incident cases of atrial fibrillation and to reduce confounding, we excluded participants who had been hospitalized before baseline evaluation with endocrine diseases or cardiovascular diseases others than hypertension, as recorded in the National Registry of Patients with the use of ICD-8 codes 240–279, 390–398, and 410–458 and ICD-10 codes E00-E90, I00-I09, and I16-I99. We did not exclude patients with hypertension who had had hypertension before or who had it at baseline for several reasons. First, the diagnostic criteria for hypertension have changed over the last decades. Second, the validity of a diagnosis of hypertension in the Danish National Register of Patients is low (15). Third, if we used a systolic blood pressure >140 mm Hg as the definition of hypertension at baseline, we would exclude >50% of subjects from the cohort (10). Finally, we did not feel a priori that the relation between blood pressure and risk of atrial fibrillation included a threshold function that allowed us to exclude any subjects from the cohort on the basis of a specific blood pressure.

The Danish Diet, Cancer, and Health Study and the present study were approved by the Regional Ethics Committees in Copenhagen and Aarhus, and by The Danish Data Protection Agency. Written informed consent was obtained from all subjects.

Baseline data
Height, weight, systolic and diastolic blood pressures, and total serum cholesterol were measured at baseline. Body weight was measured with the use of a digital scale (Soehnle-Waagen, Murrhardt, Germany) and was recorded to the nearest 100 g. Systolic and diastolic blood pressures were measured by using an automatic blood pressure measurement device (Takeda UH 751, Tokyo). Nonfasting total serum cholesterol was measured according to national guidelines (16).

All participants filled in a questionnaire about medical diseases, including myocardial infarction, angina, stroke, hypertension, hypercholesterolemia, and diabetes, and the drug treatment of those conditions. Subjects who reported that they had had ischemic heart disease, stroke, or diabetes, were medicated for those conditions, or both were excluded from the current study. The participants also completed a questionnaire about smoking habits, alcohol intake, physical activity, health, and duration of education.

Measurement of caffeine consumption
All cohort members completed a detailed, semiquantitative food-frequency questionnaire (FFQ). Descriptions of the development and validation of this questionnaire were published previously (17, 18). The daily intake of specific foods and nutrients was calculated from the FFQ for each participant with the use of FOODCALC software (version 1.3; J Lauritsen, University of Copenhagen; 19). Standard recipes and sex-specific portion sizes were applied to calculate intake in grams per day (g/d) by using data from different sources—ie, the 1995 Danish National Dietary Survey (20), 24-h diet recall interviews from 3818 of the study participants (21), and various cook books. We summed the daily intake of caffeine from coffee, tea, cola, cocoa, and chocolate for each subject.

Atrial fibrillation and atrial flutter
We identified probable cases of atrial fibrillation or flutter within the cohort in The Danish National Registry of Patients—ie, cases through 31 December 2001 with the discharge diagnoses 427.93, 427.94, and I48. A change from ICD-8 to ICD-10 occurred in Denmark in 1994. Atrial fibrillation and atrial flutter were coded separately in ICD-8, (codes 427.93 and 427.94), but, in ICD-10, they are both coded I48.

A single reviewer (LF) used a standardized form to review the medical records of the subset of study participants living in the county of Aarhus who had an incident diagnosis of atrial fibrillation or flutter recorded in the Danish National Registry of Patients through December 1999. Of 116 subjects with such a diagnosis, 112 (97% of the diagnoses) were verified by an electrocardiogram, a printout from telemetry, or a printout from a Holter recording to have atrial fibrillation or atrial flutter. Among these 112 subjects, 103 (92%) had atrial fibrillation, 3 (3%) had both atrial fibrillation and atrial flutter, and 6 (5%) had atrial flutter.

Follow-up
The study participants were followed in the National Registry of Patients and in the Danish Civil Registration System. Linking was done by use of the civil registry number. Follow-up for atrial fibrillation or flutter began on the date of visit to one of the study centers and ended on the date of an event or a censoring (ie, a diagnosis of atrial fibrillation or flutter, death, emigration, or 31 December 2001, whichever came first).

Statistical analysis
We used piecewise linear regression to examine the relation between a continuous variable and the hazard of atrial fibrillation or flutter (22). We retained the continuous variable as continuous in the Cox regression model, when appropriate, according to these analyses.

We computed a multivariate Cox proportional hazard regression model in men and women by using an initial forced entry of known risk factors for atrial fibrillation—namely, age, sex, systolic pressure, treatment for hypertension, body height, and body mass index (BMI; in kg/m²), as well as information on the quintile of caffeine consumption and the forward selection of the other variables of interest. Thereafter, we confirmed our model by doing a Cox regression analysis with backward elimination of variables. We performed supplementary analyses by adding product terms to test for interaction. The relevance of a variable in the model was further assessed by the change-in-estimate method (23).

The variables included in our final models in men and women were the quintile of caffeine consumption, age (y), sex, body height (cm), BMI, systolic blood pressure (mm Hg), treatment for hypertension (no or yes), total serum cholesterol > 6 mmol/L (no or yes), and alcohol consumption (g/d). Thereafter, we assessed further potential confounders: smoking (ie, never, former, or current) and length of education after elementary school was included with 4 categories (ie, 0 y, <3 y, 3–4 y, and >4 y) to evaluate the potential for changes in the estimate of the hazards for atrial fibrillation or flutter associated with the consumption of caffeine.

Correlation was evaluated by Spearman nonparametric correlation analysis. We used a poisson regression analysis to test for trends in incidence rates over quintiles. Tests for linear trend over quintiles were calculated by assigning the medians of intake in quintiles treated as a continuous variable.

The assumption of proportional hazards in the Cox models was evaluated by using graphic assessment and found appropriate in all models. We calculated 95% CIs throughout the analyses. All reported P values are two-sided. We used SPSS statistical software (version 11.5; SPSS Inc, Chicago) and STATA statistical software (version 7; Stata Corp, College Station, TX).

	RESULTS

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 
The cohort included 57 053 subjects at baseline. We excluded 9022 subjects who reported that they had or took medicine for endocrine or cardiovascular diseases including arrhythmias (hypertension was not excluded), had a diagnosis of endocrine or cardiovascular disease (hypertension was not excluded) in the Danish National Registry of Patients before or at baseline, or both. In addition, 42 subjects with missing questionnaires and 40 subjects with missing information on caffeine intake were excluded. Thus, the study population consisted of 47 949 subjects—22 533 men and 25 416 women.

The cohort provided a total of 275 136 person-years of risk (: 5.7 y; range: 0–8.1 y), and, during follow-up, 373 men (1.7%) and 182 women (0.7%) had an incident diagnosis of atrial fibrillation or flutter in the National Registry of Patients, which corresponded with incidence rates of 29.2 per 10 000 person-years at risk in men and 12.4 per 10 000 person-years at risk in women.

Characteristics at baseline according to quintile of caffeine consumption are shown in Table 1. The mean ± SD consumption of caffeine was 248 ± 91 mg/d in the lowest quintile and 997 ± 144 mg/d in the top quintile. Potential confounders such as age, sex, height, BMI, smoking, alcohol consumption, systolic blood pressure, and treatment for hypertension were equally distributed over quintiles of caffeine intake. The proportion of subjects with a total serum cholesterol concentration > 6 mmol/L correlated with the quintiles of caffeine intake.

Table 2

Table 3

	DISCUSSION

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

Our findings are supported both by findings from experimental studies in humans that showed no effect on the P-wave of coffee ingestion as evaluated by analysis of signal-averaged atrial electrocardiograms (24, 25) and by the finding that caffeine has no effect on interatrial and intraatrial conduction intervals, as assessed by invasive electrophysiological testing (26). Our results are also indirectly supported by previous findings that there is no risk of ventricular arrhythmias as a result of customary daily consumption of caffeine (6).

The main strengths of our study are the large number of cases with atrial fibrillation; the detailed information on potential confounding factors; the complete follow-up through nationwide, population-based registries, which limit selection and surveillance bias; and the standardized assessment of a sample of the registered outcome events. We used restriction in applying admissibility criteria, because restriction reduces confounding and increases the validity of the study (27). The incidence rates for atrial fibrillation or atrial flutter are in accordance with those for atrial fibrillation in the same age interval reported from the Framingham Heart Study (28), The Manitoba Follow-up Study (29), and the Renfrew/Paisley study (30).

The consumption of coffee in our country is high, and the range of exposure to caffeine was wide in our study. Moreover, persons who drink decaffeinated coffee may differ from those who drink ordinary coffee (31). The consumption of decaffeinated coffee in Denmark, however, is negligible: in the year 1997, consumption of decaffeinated coffee in Denmark was <1% of the total coffee consumption (32).

Our study was done in a country with a uniformly organized health care system, in which there is neither charge nor profit because the system is complete financed through taxes. Biased follow-up may have occurred, but we find it unlikely that follow-up was biased by category of caffeine consumption. We would have preferred to have a reference group that did not consume any caffeine at all. We had limited statistical power, and we were able to include only atrial fibrillation or flutter that was symptomatic and that led to hospitalization or clinical investigation in an outpatient hospital clinic. However, given the age profile of our study cohort, it is likely that patients with clinical symptoms of atrial fibrillation were referred to a hospital for further evaluation. We relied on the subjects' self-reported data on the consumption of caffeine-containing food and drink items, and we had no information on supplementary intake of caffeine from drugs. Caffeine content may vary between different brands of caffeine-containing food and drink items. Because of the observational design, we cannot exclude the possibility of unmeasured bias and uncontrolled confounding.

Some clinicians advise patients with cardiac arrhythmias to avoid caffeine-containing beverages such as coffee and tea. In this large cohort study, we could not find evidence for such a practice.

	ACKNOWLEDGMENTS

 
We thank Katja Boll, programmer, and Jytte Fogh Larsen, secretary, at The Danish Cancer Society, for assisting data collection. We thank Anne Tjønneland (Danish Cancer Society, Copenhagen) and Kim Overvad (Department of Clinical Epidemiology, Aarhus University) for providing access to the data of the Danish Diet, Cancer, and Health Study. We thank Lone Juul Hansen, statistician at The Danish Information Technology Centre for Education and Research, for assistance in data management and statistical analyses.

LF designed the study, analyzed the data, and wrote the manuscript, and PV reviewed the design, the analyses, and the manuscript. LF has received funding and grant support for research projects from Aarhus University, Boeringer-Ingelheim, Cardiome Pharma, The Danish Heart Foundation, The Danish Medical Research Council, The Danish Society of Nephrology, The Hørslev Foundation, Laerdahls Foundation for Acute Medicine, Lægekredsforeningens Forskningsfond, Merck Sharp & Dome, Nycomed, and Pfizer, and he has served as a consultant for AstraZeneca and Pfizer. PV has no personal or financial conflicts of interest.

	REFERENCES

TOP ABSTRACT INTRODUCTION SUBJECTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Benowitz NL, Osterloh J, Goldschlager N, Kaysen G, Pond S, Forhan S. Massive catecholamine release from caffeine poisoning. JAMA 1982;248:1097–8.[Medline]
2. Strubelt O, Diederich KW. Experimental treatment of the acute cardiovascular toxicity of caffeine. J Toxicol Clin Toxicol 1999;37:29–33.[Medline]
3. Chopra A, Morrison L. Resolution of caffeine-induced complex dysrhythmia with procainamide therapy. J Emerg Med 1995;13:113–7.[Medline]
4. Price KR, Fligner DJ. Treatment of caffeine toxicity with esmolol. Ann Emerg Med 1990;19:44–6.[Medline]
5. Zimmerman PM, Pulliam J, Schwengels J, MacDonald SE. Caffeine intoxication: a near fatality. Ann Emerg Med 1985;14:1227–9.[Medline]
6. Chou TM, Benowitz NL. Caffeine and coffee: effects on health and cardiovascular disease. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 1994;109:173–89.[Medline]
7. Newcombe PF, Renton KW, Rautaharju PM, Spencer CA, Montague TJ. High-dose caffeine and cardiac rate and rhythm in normal subjects. Chest 1988;94:90–4.[Abstract/Free Full Text]
8. Mehta A, Jain AC, Mehta MC, Billie M. Caffeine and cardiac arrhythmias. An experimental study in dogs with review of literature. Acta Cardiol 1997;52:273–83.[Medline]
9. Wilhelmsen L, Rosengren A, Lappas G. Hospitalizations for atrial fibrillation in the general male population: morbidity and risk factors. J Intern Med 2001;250:382–9.[Medline]
10. Tjonneland AM, Overvad OK. [Diet, cancer, and health—a population study and establishment of a biological bank in Denmark]. Ugeskr Laeger 2000;162:350–4 (in Danish).[Medline]
11. Riboli E, Hunt KJ, Slimani N, et al. European Prospective Investigation into Cancer and Nutrition (EPIC): study populations and data collection. Public Health Nutr 2002;5:1113–24.[Medline]
12. Andersen TF, Madsen M, Jorgensen J, Mellemkjoer L, Olsen JH. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull 1999;46:263–8.[Medline]
13. World Health Organization. International classification of diseases. Eighth revision (ICD-8). Danish version. Copenhagen: National Board of Health, 1986.
14. World Health Organization. International classification of diseases. Tenth revision (ICD-10). Danish version. Copenhagen: National Board of Health, 1993.
15. Nielsen HW, Tuchsen F, Jensen MV. [Validity of the diagnosis "essential hypertension" in the National Patient Registry]. Ugeskr Laeger 1996;158:163–7 (in Danish).[Medline]
16. Dyerberg J, Stender S, Andersen GE, et al. [Recommendations for clinical-chemical departments: lipid-lipoprotein analysis]. Ugeskr Laeger 1990;152:1434–7 (in Danish).[Medline]
17. Tjonneland A, Overvad K, Haraldsdottir J, Bang S, Ewertz M, Jensen OM. Validation of a semiquantitative food frequency questionnaire developed in Denmark. Int J Epidemiol 1991;20:906–12.[Abstract/Free Full Text]
18. Overvad K, Tjonneland A, Haraldsdottir J, Ewertz M, Jensen OM. Development of a semiquantitative food frequency questionnaire to assess food, energy and nutrient intake in Denmark. Int J Epidemiol 1991;20:900–5.[Abstract/Free Full Text]
19. Lauritsen J. FoodCalc software, version 1.3 (1998). Internet: http://www.foodcalc (accessed 22 July 2003).
20. Warming DL. Danskernes kostvaner, teknisk rapport 2. (Eating habits in Denmark; technical report 2.) Copenhagen: The Danish Veterinary and Food Administration, 1997 (in Danish).
21. Slimani N, Deharveng G, Charrondiere RU et al. Structure of the standardized computerized 24-h diet recall interview used as reference method in the 22 centers participating in the EPIC project. European Prospective Investigation into Cancer and Nutrition. Comput Methods Programs Biomed 1999;58:251–66.[Medline]
22. Thomsen BL. A note on the modelling of continuous covariates in Coxs regression model. Research Report 88/5. Copenhagen: University of Copenhagen Statistical Research Unit, 1988.
23. Greenland S. Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989;79:340–9.[Abstract/Free Full Text]
24. Donnerstein RL, Zhu D, Samson R, Bender AM, Goldberg SJ. Acute effects of caffeine ingestion on signal-averaged electrocardiograms. Am Heart J 1998;136:643–6.[Medline]
25. Caron MF, Song J, Ammar R, Kluger J, White CM. An evaluation of the change in electrocardiographic P-wave variables after acute caffeine ingestion in normal volunteers. J Clin Pharm Ther 2001;26:145–8.[Medline]
26. Dobmeyer DJ, Stine RA, Leier CV, Greenberg R, Schaal SF. The arrhythmogenic effects of caffeine in human beings. N Engl J Med 1983;308:814–6.[Medline]
27. Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia: Lippincott-Raven, 1998.
28. Benjamin EJ, Levy D, Vaziri SM, D'Agostino RB, Belanger AJ, Wolf PA. Independent risk factors for atrial fibrillation in a population-based cohort. The Framingham Heart Study. JAMA 1994;271:840–4.[Abstract]
29. Krahn AD, Manfreda J, Tate RB, Mathewson FA, Cuddy TE. The natural history of atrial fibrillation: incidence, risk factors, and prognosis in the Manitoba Follow-Up Study. Am J Med 1995;98:476–84.[Medline]
30. Stewart S, Hart CL, Hole DJ, McMurray JJ. Population prevalence, incidence, and predictors of atrial fibrillation in the Renfrew/Paisley study. Heart 2001;86:516–21.[Abstract/Free Full Text]
31. Kubo SA, Klatsky AL, Armstrong MA. Traits of persons who drink decaffeinated coffee. Ann Epidemiol. 2003;13:273–9.[Medline]
32. [Statistics Denmark]. Internet: http://www.dst.dk (accessed 4 August 2003) (in Danish).

This article has been cited by other articles:

				N. K Choudhry, B. Zagorski, J. Avorn, R. Levin, K. Sykora, A. Laupacis, and M. Mamdani Comparison of the Impact of the Atrial Fibrillation Follow-Up Investigation of Rhythm Management Trial on Prescribing Patterns: A Time-Series Analysis Ann. Pharmacother., November 1, 2008; 42(11): 1563 - 1572. [Abstract] [Full Text] [PDF]

				B. A. Schoonderwoerd, M. D. Smit, L. Pen, and I. C. Van Gelder New risk factors for atrial fibrillation: causes of 'not-so-lone atrial fibrillation' Europace, June 1, 2008; 10(6): 668 - 673. [Abstract] [Full Text] [PDF]

				M. G. Silletta, R. Marfisi, G. Levantesi, A. Boccanelli, C. Chieffo, M. Franzosi, E. Geraci, A. P. Maggioni, G. Nicolosi, C. Schweiger, et al. Coffee Consumption and Risk of Cardiovascular Events After Acute Myocardial Infarction: Results From the GISSI (Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico)-Prevenzione Trial Circulation, December 18, 2007; 116(25): 2944 - 2951. [Abstract] [Full Text] [PDF]

				Y.-a. Zhang, R. A. Tuft, L. M. Lifshitz, K. E. Fogarty, J. J. Singer, and H. Zou Caffeine-activated large-conductance plasma membrane cation channels in cardiac myocytes: characteristics and significance Am J Physiol Heart Circ Physiol, October 1, 2007; 293(4): H2448 - H2461. [Abstract] [Full Text] [PDF]

				Minerva BMJ, April 9, 2005; 330(7495): 854 - 854. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Frost, L. Articles by Vestergaard, P. Search for Related Content PubMed PubMed Citation Articles by Frost, L. Articles by Vestergaard, P. Agricola Articles by Frost, L. Articles by Vestergaard, P.