Skip Navigation


Rheumatology Advance Access originally published online on January 30, 2008
Rheumatology 2008 47(3):329-333; doi:10.1093/rheumatology/kem373
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
47/3/329    most recent
kem373v1
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 Disclaimer
Google Scholar
Right arrow Articles by Clarke, A. E.
Right arrow Articles by Li, T.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Clarke, A. E.
Right arrow Articles by Li, T.
Related Collections
Right arrow Systemic Lupus Erythematosus and Autoimmunity
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2008 The Author(s)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

SLE patients with renal damage incur higher health care costs

A. E. Clarke1,2, P. Panopalis3, M. Petri4, S. Manzi5,6, D. A. Isenberg7, C. Gordon8, J.-L. Senécal9, L. Joseph2,10, Y. St Pierre2 and T. Li11

1Department of Medicine, Division of Clinical Immunology/Allergy, 2Department of Medicine, Division of Clinical Epidemiology, McGill University Health Centre, McGill University, Montréal, Québec, 3Department of Medicine, Division of Rheumatology, University of California, San Francisco, CA, 4Johns Hopkins University School of Medicine, Baltimore, MD, 5Department of Medicine, Division of Rheumatology, and 6Department of Epidemiology, University of Pittsburgh, Pittsburgh, PA, USA, 7Centre for Rheumatology, Department of Medicine, University College London, London, 8Department of Rheumatology, Division of Immunity and Infection, University of Birmingham, Birmingham, UK, 9Department of Medicine, Division of Rheumatology, Centre Hospitalier de L’Université de Montréal, Université de Montréal, 10Department of Epidemiology and Biostatistics, McGill University, Montréal, Québec and 11Global Epidemiology and Outcomes Research, Bristol-Myers Squibb Company, Princeton, NJ, USA.

Correspondence to: A. Clarke, McGill University Health Centre (MUHC), 687 Pine Avenue West, V Building, Montreal, Quebec, Canada H3A 1A1. E-mail: ann.clarke{at}mcgill.ca


   Abstract
Top
 Abstract
Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
Objectives. To compare costs and quality of life (QoL) between SLE patients with and without renal damage.

Methods. Seven hundred and fifteen patients were surveyed semi-annually over 4 yrs on health care use and productivity loss and annually on QoL. Cumulative direct and indirect costs (2006 Canadian dollars) and QoL (average annual change in SF-36) were compared between patients with and without renal damage [Systemic Lupus International Collaborating Clinics/ACR Damage Index (SLICC/ACR DI)] using simultaneous regressions.

Results. At study conclusion, for patients with the renal subscale of the SLICC/ACR DI = 0 (n = 634), 1 (n = 54), 2 (n = 15) and 3 (n = 12), mean 4-yr cumulative direct costs per patient (95% CI) were $20 337 ($18 815, $21 858), $27 869 ($19 230, $36 509), $51 191 ($23 463, $78 919) and $99 544 ($57 102, $141 987), respectively. In a regression where the renal subscale of the SLICC/ACR DI was a single indicator variable, on average (95% CI), each unit increase in renal damage was associated with a 24% (15%, 33%) increase in direct costs. In a regression where each level in the renal subscale was an indicator variable, patients with end-stage renal disease incurred 103% (65%, 141%) higher direct costs than those without renal damage. Cumulative indirect costs and annual change in the SF-36 summary scores did not differ between patients.

Conclusions. SLE patients with renal damage incurred higher direct costs, but did not experience a poorer QoL. QoL may be more influenced by concurrent renal activity than accumulated renal damage, which can occur at any time and patients may gradually habituate to their compromised health state.

KEY WORDS: SLE, Direct costs, Indirect costs, Quality of life, Renal damage


   Introduction
Top
 Abstract
 Introduction
Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 
Several studies have documented the health costs [1–4] and quality of life (QoL) experienced by patients with SLE [5, 6], but few have estimated the costs [7] and QoL [8, 9] incurred specifically by SLE patients with renal involvement. Given that many patients with renal involvement receive toxic immunosuppressants and, if unsuccessful, will require dialysis or transplantation, it is anticipated that their health costs will be substantial and their well-being considerably compromised. In this study, we compared the costs and QoL between SLE patients with and without renal involvement.


   Patients and methods
Top
 Abstract
 Introduction
 Patients and methods
Results
 Discussion
 Acknowledgements
 References
 
Patients
Consecutive patients presenting to six tertiary care SLE clinics (the Montreal General Hospital and Hôpital Notre-Dame, Montreal, Quebec, Canada; Johns Hopkins University School of Medicine, Baltimore and the University of Pittsburgh, US and University College Hospital, London and the Queen Elizabeth Hospital, Birmingham, UK) and fulfilling the American College of Rheumatology (ACR) revised criteria for SLE [10] were invited to participate in a longitudinal study on health expenditure, productivity loss and QoL. Patients were enrolled between July 1995 and February 1998 and the final assessments were completed between May 1999 and October 2001. Approval was obtained from each centre's Institutional Review Board and informed consent from each participant. Cross-country comparisons of cumulative direct [1] and indirect costs [2], damage accumulation [1], and QoL [11] have been published. This report compares direct and indirect costs and QoL between patients with varying levels of renal damage.

Procedures
At study entry and semi-annually for 4 yrs, participants reported on health resource utilization, time lost from labour and non-labour market activities and time lost by their caregivers in delivering health care to the patient or helping the patient in obtaining health care (using a modified version of the Stanford Health Assessment Questionnaire [12]). At study entry and annually, patients completed questionnaires on QoL (the SF-36 [13] and a visual analogue scale adapted from the EuroQoL [14]), social support (the Interpersonal Support Evaluation List—ISEL [15]) and satisfaction with health care (the Medical Outcomes Study Patient Satisfaction Questionnaire [16]). At study entry and conclusion, treating physicians assessed disease activity [using the Systemic Lupus Activity Measure Revised (SLAM-R) [17] and a visual analogue scale of current activity and activity over the past year] and damage [using the Systemic Lupus International Collaborating Clinics/ACR Damage Index (SLICC/ACR DI) [18]].

Calculating direct and indirect costs
Given that the prices for health services and wages differ across countries, we used Canadian prices for health services and Canadian wages to value lost time so that any observed cost differences would reflect differences in the quantity of resources used or hours of productivity loss [1, 2]. Total cumulative indirect costs represent the sum of the cumulative indirect costs due to diminished labour and non-labour market activity. Non-labour market activity was valued using the replacement cost method, where estimates of the value of lost time in household work are based on the expected earnings of service workers [2]. Direct and indirect costs had previously been computed using 2002 Canadian prices. Direct costs were updated to the 2006 price level by increasing all previous costs by a factor reflecting the effect of price changes on average total costs over the full study period. A sensitivity analysis was conducted by directly applying 2006 Canadian prices to all health care goods and services used by patients during the study. However, this latter approach is only valid under the assumption that relative price changes would only have a negligible impact on resource utilization. Indirect costs were updated to 2006 using a wage index.

Statistical analyses
Demographics, QoL, direct and indirect costs and disease characteristics were compared between patients with and without renal damage using means, S.D., medians, interquartile ranges and proportions as appropriate. Renal damage was defined based on accumulation of damage in the renal items of the SLICC/ACR DI [i.e. glomerular filtration rate or creatinine clearance <50% = 1, proteinuria ≥3.5 g/day = 1 or end-stage renal disease (regardless of dialysis or transplantation)= 3]. Each item was required to be present for at least 6 months to be considered as renal damage.

For patients who provided incomplete data (i.e. those who died, withdrew, were to lost to follow-up or provided data at study entry and conclusion, but did not complete all resource use or SF-36 questionnaires), missing cost and QoL data were managed through multiple imputation using best predictive regression models with all available data from all patients as potential covariates [19]. Potential covariates included age, sex, ethnicity (Caucasian vs non-Caucasian), education (both as years and categorical as <12 or ≥12 yrs), marital status (married vs unmarried), disease duration, health status (individual SF-36 subscales, physical and mental component summary—PCS and MCS scores—and EuroQol visual analogue scale), social support (ISEL total score), patient satisfaction with health care (individual subscales), direct and indirect costs, disease activity (SLAM-R and physician-reported visual analogue scale of current activity and activity over the past year), disease damage and country delivering the health care. Consistent with our previous analyses [1, 2, 11], for participants who died during the study, imputations were performed up to 4 yrs after entry.

For the direct cost outcome, given that the error terms from linear regressions for such outcomes are often not normally distributed, a logarithmic transformation was performed [20]. Since the disease course fluctuates, to better characterize the long-term change in QoL, all SF-36 subscale and summary scores were used to estimate a linear trend across the entire study interval within each individual patient. This was accomplished through two-level hierarchical linear modelling, an approach that allows for the borrowing of strength across patients, while allowing for individual within-patient variations.

Direct and indirect costs and the patient-specific rate of change in the SF-36 PCS and MCS scores were then compared between patients with varying levels of renal damage using simultaneous regressions. Simultaneous regression equations are related only through the correlation of their error terms that can help minimize the variance around the estimated coefficients [21]. Only study entry values of the previously mentioned covariates were considered. For each outcome (i.e. direct and indirect costs and rate of change in PCS and MCS scores), two regression models are presented, one where the renal subscale of the SLICC/ACR DI was considered as a single covariate and one where each level in the SLICC/ACR DI was considered as a single covariate.

Model selection for all regressions was based on the Bayes factor as approximated by the Bayesian Information Criteria.


   Results
Top
 Abstract
 Introduction
 Patients and methods
 Results
Discussion
 Acknowledgements
 References
 
Seven hundred and fifteen patients were enrolled (Canada 231, US 269, UK 215) (Table 1). Sixty-eight per cent of patients with a renal subscale score of 0, 65% with a score of 1 and 67% with a score of 2 or 3 completed at least four of a maximum of seven resource utilization/productivity loss questionnaires, and at least three of five SF-36 questionnaires. Five per cent of patients with a renal subscale score of 0, 11% with a score of 1, 0% with a score of 2 and 17% with a score of 3 died. Twenty-seven per cent of patients with a renal subscale score of 0, 24% with a score of 1, 34% with a score of 2 and 17% with a score of 3 provided incomplete data.


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

  		TABLE 1. Baseline characteristics of study participants

 
Patients with a renal subscale score of 3 were less likely to be Caucasian and had longer disease duration (Table 1). The PCS scores and disease activity did not differ meaningfully between patient groups. The percentage of patients employed at baseline as well as the baseline SF-36 subscale scores did not differ between patients with and without renal damage (data not shown).

When all patients were included by using multiple imputation for those who were deceased or provided incomplete data, the mean (95% CI) 4-yr cumulative direct medical costs were $20 337 ($18 815, $21 858), $27 869 ($19 230, $36 509), $51 191 ($23 463, $78 919) and $99 544 ($57 102, $141 987), respectively for those with a SLICC renal subscale score = 0 through 3 (Table 2). For patients with a renal subscale score of 3, the largest components of total costs were hospitalizations ($22 356) and dialysis ($58 013). The 4-yr cumulative indirect costs did not differ appreciably between patient groups (Table 2). The mean annual change in the PCS and MCS scores was clinically negligible and did not differ between those with and without renal damage.


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

  		TABLE 2. Patient characteristics at final study visita

 
In the regression model for cumulative direct costs (Table 3), the renal damage coefficient represents the percentage change in cumulative costs relative to a renal damage score of zero. In the regression model where the renal subscale of the SLICC/ACR DI was a single indicator variable, on average (95% CI), each unit increase in renal damage was associated with a 24% (15%, 33%) increase in costs. In the model where each level in the renal subscale was an indicator variable, patients with a decreased glomerular filtration rate and proteinuria incurred 43% (11%, 76%) higher costs than those without renal damage and those with end-stage renal disease incurred 103% (65%, 141%) higher costs than those without renal damage. The cumulative indirect costs and annual change in the SF-36 PCS and MCS scores did not differ between patients with and without renal damage.


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

  		TABLE 3. Regression model for cumulative direct and indirect medical costs and average annual change in SF-36 physical and mental summary scores

 
Results (data not shown) were not substantially different when using the alternative costing method of applying 2006 Canadian prices directly to each health care item rather than a unique price index.


   Discussion
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
Acknowledgements
 References
 
We have shown that SLE patients with renal damage incurred substantially higher direct costs than those without renal damage, but did not experience less work productivity. Further, although SLE patients with and without renal damage experienced a poorer QoL than the general population, annual change in QoL did not differ between SLE patient groups. It is possible, however, that if we had a larger sample of patients with renal damage, we would have observed less productivity and an inferior QoL in this group. Although the SF-36 is a widely used and well validated generic health status instrument that allows comparison across a variety of conditions, it is insensitive to several SLE-specific manifestations [5]. Hence, SLE-specific instruments that characterize health status domains particularly important for patients with SLE, such as body image, self-confidence, reproductive ability and support and attitudes of family and friends [22], may have greater sensitivity to changes in renal damage than the SF-36.

It is probable that QoL (as assessed by the SF-36) is not substantially influenced by accumulated renal damage. Renal damage may occur at any time during the disease course and, although irreversible, patients may gradually habituate to their compromised health state. Further, with the advent of transplantation, renal damage may be less likely to affect QoL than other facets of lupus. It is possible that QoL is more influenced by concurrent renal activity and non-renal lupus manifestations that cause pain. However, in exploratory analyses, we were unable to demonstrate that patients undergoing dialysis or receiving cyclophosphamide experienced a poorer QoL (data not shown). Although few studies have examined the influence of renal damage or renal activity specifically on QoL, it has been shown that neither overall disease damage nor activity correlates well with QoL and that these are largely independent constructs [6].

Only two other studies have specifically assessed QoL of SLE patients with renal involvement [8, 9]. Vu and Escalante [9] compared QoL, as expressed on the SF-36, between 22 patients undergoing maintenance renal dialysis and 82 patients with preserved renal function. Patients on dialysis had poorer physical function and general health than those with preserved renal function, but, surprisingly, had superior mental health. Tse et al. [8] assessed QoL in 12 SLE patients during their first 6 months of therapy with either cyclophosphamide or mycophenolate mofetil for proliferative nephritis. Although the rate of complete remission was identical for both the therapies (83.3%), mycophenolate was associated with a superior QoL, likely due to its reduced side-effect profile.

These studies on QoL in SLE patients with renal involvement are limited. They are short term and likely misrepresent how patients live with lupus over time. Systemic lupus, punctuated by episodes of exacerbation and remission, results in fluctuating levels of health status and longitudinal studies are necessary to fully characterize the influence of the disease on patients’ lives. Further, the Tse study relied on patients’ recall of remote experiences [8] and did not include a comparator group without renal involvement; the Vu study [9] only involved patients with the most severe level of renal dysfunction. Our study evaluated QoL using longitudinal data provided by patients spanning the full spectrum of renal dysfunction.

It remains to be seen how future treatments will alter costs and health status in SLE. Thus far, medications have not contributed to the total cost of SLE to the same degree as they have in other inflammatory rheumatic conditions such as rheumatoid arthritis [3, 23]. Nevertheless, improved understanding of the pathogenesis of SLE combined with advances in biotechnology have led to the rapid emergence of many potential treatments which, by more specific targeting of the immune system, should be more efficacious with less toxicity [24]. However, their costs are likely to be substantial. Given the considerable direct costs in SLE associated with renal replacement, the even greater costs due to lost productivity, and the intangible costs of infertility and an otherwise compromised QoL, it is hoped that the anticipated benefits of these novel therapeutics will be commensurate with their costs.

Formula


   Acknowledgements
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
References
 
This authors would like to acknowledge the support of Tri-nation Staff: Montreal General Hospital: Tina Panaritis BA, Popi Panaritis, Kathy Margonis, Maura Trifero RN, Diane Ferland RN, Carolyn Neville BA, RN, Mary Orsini-Dudin; University of Pittsburgh: Joan Rairie RN, BSN; University of Birmingham: Stephanie Heaton RGN, Lupus UK Specialist Research Nurse. The authors thank all the participating physicians (University of Birmingham: Margaret Allen PhD, MD; Simon Bowman PhD, MD), Kim Allan for her expert technical assistance, and the patients, which made this study possible.

Funding: Supported by grants from Bristol-Myers Squibb Company, the Fonds de la recherche en santé du Québec, and The Arthritis Society of Canada. The Montreal General Lupus Cohort is partially supported by the Singer Family Fund for Lupus Research; the Hopkins Lupus Cohort is supported by National Institutes of Health (NIH) RO1 AR43727 and by the Outpatient Clinical Research Center, RR 00722; the Pittsburgh Cohort is supported by the Lupus Foundation, Pennsylvania Chapter, K24 AR00213, NIH RO1 AR46588, Arthritis Foundation, National, NIH/5RO1 AL54900-02; the Birmingham Cohort is supported in part by the Wellcome Trust Clinical Research Facility and Lupus UK. Dr Clarke and Dr Joseph are National Research Scholars of the Fonds de la recherche en santé du Québec; Dr Panopalis is a Research Fellow of the Canadian Institutes for Health Research; Dr Petri is supported by the Hopkins Lupus Cohort RO1 AR43727-06 and the General Clinical Research Center MO1-RR 00052; Dr Li is an employee of Bristol-Myers Squibb Company.

Disclosure statement: A.C. is a consultant for Human Genome Sciences and Bristol-Myers Squibb and has been or is a site investigator for Human Genome Sciences and Med Immune. C.G. has acted to Bristol-Myers Squibb as a consultant on a clinical trial in systemic lupus erythematosus. S.M. serves as a member of the advisory Board for Bristol-Myers Squibb. T.L. is an employee of Bristol-Myers Squibb and owns company stocks. All other authors have no conflicts of interest.


   References
Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Clarke AE, Petri M, Manzi S, et al. The SLE Tri-Nation Study: absence of a link between health resource use and heath outcome. Rheumatology (2004) 43:1016–24.[Abstract/Free Full Text]
  2. Panopalis P, Petri M, Manzi S, et al. The Systemic Lupus Erythematosus Tri-nation Study: cumulative indirect costs. Arthritis Rheum (2007) 57:64–70.[CrossRef][ISI][Medline]
  3. Huscher D, Merkesdal S, Thiele K, Zeidler H, Schneider M, Zink A. Cost of illness in rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis and systemic lupus erythematosus in Germany. Ann Rheum Dis (2006) 65:1175–83.[Abstract/Free Full Text]
  4. Nichol MB, Shi S, Knight TK, Wallace DJ, Weisman MH. Eligibility, utilization, and costs in a California Medicaid lupus population. Arthritis Rheum (2004) 51:996–1003.[CrossRef][ISI][Medline]
  5. Thumboo J, Strand V. Health-related quality of life in patients with systemic lupus erythematosus: an update. Ann Acad Med Singapore (2007) 36:115–22.[ISI][Medline]
  6. McElhone K, Abbott J, Teh LS. A review of health related quality of life in systemic lupus erythematosus. Lupus (2006) 15:633–43.[Abstract/Free Full Text]
  7. McInnes PM, Schuttinga J, Sanslone WR, Stark SP, Klippel JH. The economic impact of treatment of severe lupus nephritis with prednisone and intravenous cyclophosphamide. Arthritis Rheum (1994) 37:1000–6.[ISI][Medline]
  8. Tse KC, Tang CS, Lio WI, Lam MF, Chan TM. Quality of life comparison between corticosteroid- and-mycofenolate mofetil and corticosteroid- and-oral cyclophosphamide in the treatment of severe lupus nephritis. Lupus (2006) 15:371–9.[Abstract/Free Full Text]
  9. Vu TV, Escalante A. A comparison of the quality of life of patients with systemic lupus erythematosus with and without end stage renal disease. J Rheumatol (1999) 26:2595–601.[ISI][Medline]
  10. Hochberg MC. Updating the American college of rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum (1997) 40:1725.[ISI][Medline]
  11. Panopalis P, Petri M, Manzi S, et al. The Systemic Lupus Erythematosus Tri-nation Study: longitudinal changes in physical and mental well-being. Rheumatology (2005) 44:751–5.[Abstract/Free Full Text]
  12. Fries JF, Spitz PW, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum (1980) 23:127–34.[ISI][Medline]
  13. Ware JE Jr, Sherbourne CD. The MOS 36 item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care (1992) 30:473–83.[ISI][Medline]
  14. The EuroQol Group. EuroQol: a new facility for the measurement of health-related quality of life. Health Policy (1990) 16:199–208.[CrossRef][ISI][Medline]
  15. Cohen AS, Mermelstein R, Kamarck T, et al. Measuring the functional components of social support. In: Sarason IG, Sarason BR, eds. Social support: theory, research, and applications. The Hague: Martinus Nijhoff (1985) 73–94.
  16. Ware JE Jr, Snyder MK, Wright R, Davies AR. Defining and measuring patient satisfaction with medical care. Eval Program Plann (1983) 6:247–63.[CrossRef][Medline]
  17. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang C. The Committee on prognosis studies in SLE. Derivation of the SLEDAI. A disease activity index for lupus patients. Arthritis Rheum (1992) 35:630–40.[ISI][Medline]
  18. Gladman DD, Ginzler EM, Goldsmith CH, et al. The development and initial validation of the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index for Systemic Lupus Erythematosus. Arthritis Rheum (1996) 39:363–9.[ISI][Medline]
  19. Rubin D. Multiple imputation for nonresponse in surveys (1987) New York: Wiley.
  20. Aitchison J, Brown JAC. The lognormal distribution: with special reference to its use in economics (1957) Cambridge: Cambridge University Press.
  21. Greene WH. Systems of regression equations. Ch. 17. Econometric analysis. (1993) 2nd. New York: Macmillan.
  22. Stamm TA, Bauernfeind B, Coenen M, et al. Concepts important to persons with systemic lupus erythematosus and their coverage by standard measures of disease activity and health status. Arthritis Rheum (2007) 57:1287–95.[CrossRef][ISI][Medline]
  23. Rat AC, Boissier MC. Rheumatoid arthritis: direct and indirect costs. Joint Bone Spine (2004) 71:518–24.[CrossRef][ISI][Medline]
  24. Isenberg D, Rahman A. Systemic lupus erythematosus—2005 annus mirabilis? Nat Clin Pract Rheumatol (2006) 2:145–52.[CrossRef][ISI][Medline]
Submitted 28 August 2007; revised version accepted 17 December 2007.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
47/3/329    most recent
kem373v1
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 Disclaimer
Google Scholar
Right arrow Articles by Clarke, A. E.
Right arrow Articles by Li, T.
PubMed
Right arrow PubMed Citation
Right arrow Articles by Clarke, A. E.
Right arrow Articles by Li, T.
Related Collections
Right arrow Systemic Lupus Erythematosus and Autoimmunity
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?