Skip Navigation

European Journal of Heart Failure 2000 2(3):291-297; doi:10.1016/S1388-9842(00)00089-1
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (16)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Linné, A. B.
Right arrow Articles by Israelsson, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Linné, A. B.
Right arrow Articles by Israelsson, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2000 European Society of Cardiology

Health care costs of heart failure: results from a randomised study of patient education

A. Björck Linnéa,*, H. Liedholma, S. Jendtegb and B. Israelssonc

a Department of Community Medicine, Malmö University Hospital, University of Lund Sweden
b The Swedish Institute of Health Economics Lund, Sweden
c Department of Cardiology, Malmö University Hospital, University of Lund Sweden

* Corresponding author. Department of Community Medicine, MFC, Malmö University Hospital, S-205 02 Malmö, Sweden. Tel.: +46-40-33-34-49; fax: +46-40-33-62-15. E-mail address: agneta.bjork-linne{at}smi.mas.lu.se (A. Björck Linné).


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Background: Heart failure is a serious syndrome with a bad prognosis. Hospitalisation is common and readmittance rate is high; factors which influence the cost of care and treatment. Only scarce data on detailed patient materials regarding health care costs are known.

Aims: To describe in detail the health care costs for heart failure patients.

Methods: Costs for patients (n = 108) who completed a randomised education trial were studied for 6 months after hospital discharge. Costs for hospital stay, out-patient visits, diagnostic tests and procedures, laboratory analyses and drug treatment were calculated. Official unit prices list used to reimburse providers of cross-boundary health services and prices for drugs in the Swedish Drug Compendium were employed.

Results: The total cost for a heart failure patient was approximately 20000 SEK (2564 US$, 7.80 SEK = 1 US$) for 6 months. There was a 27-fold variation between patients. There was no relation between age or sex and cost. In decreasing order cost for hospitalisation was followed by costs for out-patient visits, diagnostic tests and procedures, laboratory analyses and drugs.

Conclusion: Hospitalisation was the largest part of the total cost and there was a large inter-individual variation. Efforts to reduce the economic burden should be focused on hospitalisation. Due to skewed distribution, individual data must be considered in the analysis of the efforts.

Key Words: Heart failure • Health care costs • Individual patients

Received January 14, 2000; Revised March 20, 2000; Accepted May 12, 2000


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
The syndrome of heart failure is an increasing health care problem, mostly due to a growing proportion of older persons in the population and to improved survival after myocardial infarction. The prevalence of heart failure in Sweden is estimated to be 2–3% for the total population and 10–15% for persons >75 years old [1]. The treatment of heart failure has been improved, much due to the introduction of ACE-inhibitors [2]. Still, heart failure is a serious syndrome with a bad prognosis, hospitalisation is common and readmittance rate is high [3]. These factors highly influence the cost of care and treatment of the heart failure patients [4–8]. Much effort has been spent to lessen readmittance rate and minimise hospital stay of patients in order to decrease the economic burden [5,7,9,10].

During 1995–1997 we performed a randomised, controlled trial on heart failure patients in Sweden in order to measure the impact of an interactive patient education programme added to standardised information on patients’ knowledge of their disease and treatment [11]. The study showed increased knowledge of items of importance for self management of heart failure in the intervention group compared to the control group. The results were valid 6 months after education. According to the protocol of the study, data on hospitalisation, out-patient visits, investigations and medical treatment were collected in order to describe the health care costs. Although the total cost for heart failure treatment is high, relatively few studies describe details of the costs. The aim of this report is to describe these on an individual level.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Patients
The study population consisted of 108 patients (68 men and 40 women), aged 42–85, who completed the education trial, which originally consisted of 130 randomised patients.

Patients, irrespective of age, treated for heart failure at the heart failure unit as in-patients and discharged to a control visit to a specialist nurse, were eligible for this education trial. Criteria for heart failure were left ventricular ejection fraction (LVEF)<40% (echocardiography) or at least two of the following criteria: pulmonary rales, peripheral oedema, a third heart sound and signs of heart failure at chest X-ray. Patients with somatic disease or physical handicap with difficulties to communicate or handle the technical equipment, expected problems with compliance due to alcohol/drug abuse or major psychiatric illness, and participation in a drug trial were excluded. Participant flow of the study has been presented recently [11]. Patient characteristics at randomisation are presented in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1 Patient characteristics at randomisationa

 
Since the current study was not designed to study the differences in health care costs between the two groups, patients from the intervention group (n=50) and the control group (n=58) are presented together. Patients were studied from the time of randomisation, 7–14 days after discharge from hospital, for a period of 6 months.

2.2. Use of health care resources
Costs for hospital stay, out-patient visits, investigations, laboratory analyses and drug treatment for the study period of 6 months were calculated. Except for drug treatment, costs were calculated from the official 1998 unit prices list used to reimburse providers of cross-boundary health services (The Regional Joint Committee of the Southern Sweden Health Care Region). Price lists included costs for in-patient care, out-patient care, diagnostic tests and procedures and laboratory tests.

2.3. Hospital stay
Information on hospital stay at the University Hospital as well as the Geriatric Rehabilitation Hospital was collected from the computerised patient administrative system (PAS). This system contains information on community-based health care [e.g. hospital stay and visits to general practitioners (GP)] and is continually updated. In this study, all causes for hospitalisation were counted. Information from the Departments of Alcohol and Drug Diseases, Psychiatry, and Obstetrics and Gynaeocology is not accessible to other health-care providers for integrity reasons. However, the omission of this information is regarded to be of limited importance for the results.

2.4. Out-patient visits
Information on visits was taken from reports made by the patients at interviews (at 1 month and 6 months after education), tracked and verified by the PAS. Exceptions were only out-patient visits at above-mentioned clinics. Personal letters to private caregivers, whose care is not covered by the PAS, were sent for verification of attendance. Visits to cardiologists and general practitioners were counted and classified irrespective of the type of the doctor’s employment. Attendance to specialist nurses on an out-patient basis were taken from log lists at the Heart Failure Clinic. Attendance to primary care nurses were, however, not included.

2.5. Diagnostic tests and procedures
Information on diagnostic tests and procedures performed during the study period was collected from patient records. The procedures covered electrocardiograms, X-ray investigations (including angiograms), scintigrams, echocardiograms, ergometer tests, spirometries, etc. Procedures such as heart catheterisation and PTCA were also included, although seldom practised.

2.6. Laboratory analyses
Information on analyses performed during the study period was also collected from patient records. Laboratory analyses covered blood and urine chemistry, etc.

Diagnostic tests and procedures and laboratory analyses initiated at visits to other clinics and health care centres than the medical clinics at the hospital were not included because of problems with data retrieval.

2.7. Medication
Information on actual drug treatment at the time of randomisation, the 1-month test and the 6-month test were collected from individual medication lists. These are originally given to the patients at discharge from the hospital and changed as needed by health care personnel during follow up. As an initial screening of the medication lists gave the impression of very small intra-individual changes during the study period, we employed a sampling method, which yielded data from 15 randomly selected patients. The analysis confirmed the observation of low intra-individual variation of treatment. For simplification, medication lists at 6 months were then used for calculation of drug treatment costs. Prices for the drugs were taken from FASS (The Swedish Drug Compendium), cash values valid from 1 January 1998.

2.8. Statistics
Conventional descriptive parametric and non-parametric statistical methods were used. Arithmetic mean, standard deviation (S.D.), 95% confidence interval (CI), median and range and linear regression were used when appropriate.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
The total cost (median) for a single heart failure patient was approximately 20 000 SEK (2564 US$, 7.80 SEK=1 US$) for a 6-month period (Table 2). However, there was a large variation between patients (Fig. 1) and the cost was not related to age (Fig. 2) or sex (data not shown). The contribution of an individual cost item can highly influence the total sum. The cost for 1 day’s stay at the department of cardiology amounts to 2798 SEK; a performance of PTCA, 23 792 SEK; a visit to a GP, 630 SEK; and an ECG, 192 SEK. The contribution of costs for laboratory analyses was very small as were the costs for drugs for heart failure treatment. Of the drug cost, 65% was utilised for ACE-inhibitors, 18% for diuretics, 8% for nitrates, 4% for potassium sparing drugs, 3% for digitalis and 2% for aspirin. There was no relation between age and cost for drugs (Fig. 3).


View this table:
[in this window]
[in a new window]

  		Table 2 Health care costs per patient (SEK) during 6 months of follow-up after hospitalisation for heart failurea

 


Figure 1
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Ranking of individual patients (n=108) after total cost. Data are individual patients (x-axis) and cost in SEK (1998 prices) during 6 months (y-axis).

 


Figure 2
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Relation between age of individual patients and total cost (SEK 1998 prices). Line of regression and 95% CI is indicated.

 


Figure 3
View larger version (12K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Relation between age of individual patients and cost of drugs (SEK 1998 prices) for treatment of heart failure. Line of regression and 95% CI is indicated.

 
In order to give examples of some clinical characteristics of patients in the investigation, two patients from each study group were chosen from the extremes of the total cost distribution at 6 months.

Woman, aged 77. She was hospitalised 3 days for heart failure before discharge and the recruitment to the study. Main diagnoses were heart failure and hypertension. The patient was in NYHA-class II and her serum creatinine was 90 µmol/l at the beginning of the study. She was treated with furosemide 40 mg/day and enalapril 20 mg/day. No readmittance was necessary during the 6-month follow-up period, during which the patient visited a specialist nurse once and a GP once. The cost for heart failure drugs was 1307 SEK and the total cost (drugs included) for 6 months was 4615 SEK.

Man, aged 83. He was hospitalised 6 days for heart failure before discharge and recruitment to the study. Main diagnoses were heart failure and prior myocardial infarction. The patient was in NYHA-class II and his serum creatinine was 128 µmol/l. The patient’s treatment was furosemide 40 mg/day, enalapril 10 mg/day and digoxin 0.13 mg/day. No readmittance was necessary during the 6 months follow-up. During that time the patient visited a specialist nurse four times. The cost for heart failure drugs was 1056 SEK and the total cost for 6 months was 4605 SEK.

Man, aged 59. He was hospitalised 5 days for heart failure before discharge and recruitment to the study. Main diagnoses were heart failure, atrial fibrillation, angina pectoris and obstructive lung disease. The patient was in NYHA-class III and his serum creatinine was 85 µmol/l. He was treated with furosemide 160 mg/day, digoxin 0.25 mg/day, amiloride 5 mg/day and isosorbide mononitrate (slow release) 30 mg/day. The patient was readmitted after 130 days. During the study period the patient was hospitalised six times and visited a specialist nurse four times, a cardiologist three times, a GP four times and the emergency ward three times. The cost for heart failure drugs was 796 SEK and the total cost for 6 months was 115 925 SEK.

Man, aged 71. He was hospitalised 5 days for heart failure before discharge and recruitment to the study. Main diagnoses were heart failure, atrial fibrillation, hypertension and diabetes. The patient was in NYHA-class II and his serum creatinine was 101 µmol/l. This man was treated with furosemide 80 mg/day, enalapril 20 mg/day, digoxin 0.25 mg/day and aspirin 75 mg/day. The patient was readmitted after 71 days. During the study period the patient was hospitalised five times and visited a specialist nurse five times, a GP twice and the emergency ward twice. The cost for heart failure drugs was 1531 SEK and the total cost for 6 months was 124 997 SEK.

During the study 45 of 108 patients were hospitalised (all causes). Among those hospitalised, the number of hospital visits ranged from one to six (median: two visits) and the number of days (median) spent in hospital were 10 days. The consumption of health-care services, expressed as number of visits to the emergency ward, cardiologist, GP and specialist nurse (at the ward), are shown in Table 3. Again, there were large variations between patients (Fig. 4) and there was no relation between age and use of health-care facilities (Fig. 5).


View this table:
[in this window]
[in a new window]

  		Table 3 Use of health care resources during 6 months of follow up after hospitalisation for heart failurea

 


Figure 4
View larger version (57K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Ranking of individual patients after number of visits to health-care providers. Data are individual patients (x-axis) from control group (n=58) and number of visits during 6 months (y-axis).

 


Figure 5
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Relation between age of individual patients and number of visits to health care providers. Line of regression and 95% CI is indicated.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
We calculated the cost for health care of heart failure patients in Sweden, based on randomised patients who completed a trial of patient education. As expected, the largest part of the cost was due to hospitalisation of the patients. The second largest cost was out-patient visits followed by cost for diagnostic tests and procedures, with a great inter-individual variation. In comparison, the cost for laboratory analyses and drugs was small. There was a 27-fold variation in total cost ranging from 4600 to 125 000 SEK. The large variation was seen in all items of cost and the data were not normally distributed.

Since there is only one hospital in the catchment area it is conceivable that almost all relevant events have been recorded. As we did not register diagnostic tests and procedures initiated outside the hospital, there is an underestimation of this part of the total cost. However, this underestimation is probably small as most of the costly activities are generated during the care of hospitalised patients. Participating patients were regarded to be representative of the general heart failure patients in the catchment area.

As this study only concerned the health care costs of heart failure, we did not estimate indirect costs due to time off from work or other activities during hospital stay, out-patient visits and sick leave. In principle, time loss is a cost to society due to lost productivity [8]. Furthermore, the influence of randomisation effects on the intervention group was not studied. This was regarded to be low as the intervention group was only seen two times more for education than the control group at the beginning of the study. These factors should have been included if this was a full analysis of the costs.

In reviews over health-care costs for heart failure the importance of lessening the readmittance rate to minimise the costly hospital stay has been stressed [6,7]. McMurray and Davie estimated that the need for hospitalisation could be reduced by approximately 30% if the results of trials on ACE-inhibitors for heart failure were implemented [12]. The cost of heart failure was illustrated by comparisons of calculations from different countries (The UK, Netherlands and USA). As data were derived from countries with very different health-care systems, they were not easy to compare, but it was concluded that hospitalisation was the largest cost.

Calculations of costs can either be based on facts accumulated in national registers and data bases or data generated by formal studies. Rydén-Bergsten and Andersson estimated the 1996 average treatment cost for heart failure in Sweden to be 25 000 SEK (exchange rate 6.80 SEK=1 US$, December 1996) per patient and year, based on official statistics and various clinical and epidemiological studies [4]. The calculations covered institutional care including costs for nursing home, out-patient care, surgery and drugs. Institutional care was the largest part of the cost. Cost for drugs for treatment of heart failure and cardiovascular disorders was 11% of the total cost. Only 36% of the patients were treated with ACE-inhibitors. Rydén-Bergsten and Andersson concluded that optimal use of management programs and ACE-inhibitors are important to reduce the costs.

There are few papers published on actual health-care costs of heart failure patients. Recently Stewart et al. presented the medical and health economic impact of a home-based intervention program in a randomised, controlled trial [10]. In brief, the intervention significantly decreased the rate of unplanned readmissions during 6 months, but there was no difference in hospital-based costs, largely dependent on the great inherent skewed distribution.

Two studies on Swedish patient materials have been published [5,8]. The study by Cline et al. was a study from the same department as ours and was performed 5 years earlier. This was a study of effects of a management programme on hospitalisation and health-care cost 1 year after the admission for heart failure. The cost for hospitalisation and out-patient visits was reported. The calculation of hospitalisation was based on an unofficial price list, whereas our calculation was based on the official price list for health care in southern Sweden. The mean annual health-care cost for patients (intervention group) was 2294 US$ (exchange rate 1 US$=7.76 SEK), 18 000 SEK. Our estimation was 20 000 (median) SEK for 6 months. The costs in the two studies were based on different time periods and study duration (1 year and 6 months, respectively) and as there is no linear relation between readmittance or hospitalisation and time, comparison is difficult to make.

There were also differences between these studies regarding the use of ACE inhibitors. In our study 90% were treated with ACE-inhibitors, whereas in the study by Cline et al. on average 22% were treated in the beginning and 64% at the end. Despite the high use of ACE-inhibitors in our study, the mean readmittance rate was 42% over 6 months. This is not far from the estimated figure of events in the study by Cline et al. In the ongoing discussion about benefits of ACE-inhibitor treatment these two studies show that use of ACE-inhibitors is only one part of the efforts to decrease readmittance.

In the study by Zethraeus et al., costs for 25 patients with heart failure were studied 1 year after hospitalisation and compared to corresponding, retrospective data 1 year before [8]. The hospitalisation costs were based on DRG costs (DRG=Diagnosis Related Group) for in-patient care of heart failure patients. The cost per day at the Department of Medicine (Södertälje Hospital) was 3111 SEK, which is not far from 2798 SEK used in our calculations for patients at the Department of Cardiology (Malmö University Hospital). The mean cost for heart failure was estimated to be 40 791 SEK for 1 year. In our study the median total cost was 19 682 SEK (mean 34 165) for 6 months. However, as discussed above, costs over different lengths of time cannot easily be compared.

From the studies discussed above it is evident that distributions of the usage of health-care facilities (and costs) are skewed and averages must therefore be presented as medians instead of means. This was also recently noted by Stewart et al. [10]. Analysing data on cost for heart failure care can be done in two principally different ways with different external validity. When data are calculated ‘top down’, using data from official statistics and medical publications, the result will often be presented as averages, with no view on variation [4]. When data are presented from actual patient materials, as in our study (‘bottom up’), they can also include the variation between patients. However, individually derived data depend very much on the way patients are recruited.

Accurate comparison of studies can only be made if the components are well defined, e.g. study material (in- and out-patients), health-care organisation, severity of disease, medical treatment and start and duration of study. The organisation of the care of the heart failure patients has also to be defined. Due to the great inter-individual variation in costs, a randomised, controlled trial of any intervention has to be very large to show a significant health–economic impact. In our study there was no statistical difference between the intervention and control groups regarding total cost (data not shown) and the power (post hoc) to detect a real difference did not exceed 30%.

In the beginning of the disease, maybe illustrated by patients in the low cost end of our patient distribution, few resources are needed. At the end of the disease, maybe illustrated by patients in the high cost end of the distribution, all possible resources have already been in use. For patients with advanced disease readmittance and longer periods of hospitalisation can not be avoided.

New principles for treatment affecting morbidity and mortality have emerged. Implementation of results of recent studies on beta-blockers for heart failure, e.g. CIBIS II [13], would — if they are shown to be valid on a more general basis — diminish the cost for hospitalisation to a corresponding low drug cost. A target dose of 10 mg bisoprolol would only increase the drug cost with 630 SEK per treated patient for 6 months.

In conclusion, we have confirmed previous reports that the cost for hospitalisation is the largest part of the total cost of health care for heart failure patients. Low costs for diagnostic tests and procedures and drugs makes it obvious that efforts to decrease the economic burden should be focused on hospitalisation. In our study there was a large inter-individual variation not related to age, reflecting the various needs of individual patients and hence the need for a flexible management system. The flexibility was illustrated by the role played by the specialist nurses as reflected in the number of visits. The variation in needs for resources has great impact on the presentation and comparison of studies on heart failure costs. In order to show significant effects on readmittance rate and cost, large-scale studies using strict methodological criteria have to be performed to achieve high internal and external validity.


   Acknowledgements
 
This work was supported by grants from the Swedish National Corporation of Pharmacies, the Council for Medical Health Care Research in South Sweden, the Ernhold Lundström’s Fund and the Swedish Heart and Lung Foundation. We wish to acknowledge the valuable secretarial help from Mrs Janet Feldtblad and the co-operation of private physicians in Malmö.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Information från Läkemedelsverket. Behandling av akut och kronisk hjärtsvikt. Rekommendationer. (Treatment of acute and chronic heart failure. Recommendations), no. 3, Uppsala, Sweden: Läkemedelsverket, 1996.
  2. Consensus recommendations for the management of chronic heart failure 1999. Am J Cardiol 1999;83:1A–38A.
  3. Gooding J., Jette A.M. Hospital readmissions among the elderly. J Am Geriatr Soc (1985) 33:595–601.[Web of Science][Medline]
  4. Rydén-Bergsten T., Andersson F. The health care costs of heart failure in Sweden. J Intern Med (1999) 246:275–284.[CrossRef][Web of Science][Medline]
  5. Cline C., Israelsson B., Willenheimer R., Broms K., Erhardt L. Cost effective management programme for heart failure reduces hospitalisation. Heart (1998) 80:442–446.[Abstract/Free Full Text]
  6. Rich M.W., Nease R.F. Cost-effectiveness analysis in clinical practice. The case of heart failure. Arch Intern Med (1999) 159:1690–1700.[Abstract/Free Full Text]
  7. Parmley W. Cost-effective management of heart failure. Clin Cardiol (1996) 19:240–242.[Web of Science][Medline]
  8. Zethraeus N., Molin T., Henriksson P., Jönsson B. Costs of coronary heart disease and stroke: the case of Sweden. J Intern Med (1999) 246:151–159.[CrossRef][Web of Science][Medline]
  9. Rich M., Beckman V., Wittenberg C., Leven C., Freedland K., Carney R. A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure. New Engl J Med (1995) 333:1190–1195.[Abstract/Free Full Text]
  10. Stewart S., Marley J., Horowitz J. Effects of a multidisciplinary, home-based intervention on unplanned readmissions and survival among patients with chronic congestive heart failure: a randomised controlled study. Lancet (1999) 354:1077–1083.[CrossRef][Web of Science][Medline]
  11. Björck Linné A., Liedholm H., Israelsson B. Effects of systematic education on heart failure patients’ knowledge after 6 months. A randomised, controlled trial. Eur J Heart Failure (1999) 1:219–227.[Abstract/Free Full Text]
  12. McMurray J., Davie A. The pharmacoeconomics of ACE inhibitors in chronic heart failure. PharmacoEconomics (1996) 9:188–197.[Web of Science][Medline]
  13. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353(9146):9–13.

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur J Heart FailHome page
C. Berry, D. R. Murdoch, and J. J.V. McMurray
Economics of chronic heart failure
Eur J Heart Fail, June 1, 2001; 3(3): 283 - 291.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (16)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Linné, A. B.
Right arrow Articles by Israelsson, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Linné, A. B.
Right arrow Articles by Israelsson, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?